A QUASI-EXPERIMENTAL STUDY OF AN INTERVENTION USING THE TEXT

STRUCTURE STRATEGY IN GRADE 7 SCIENCE

A Dissertation

by

ANDREA LYNNE BEERWINKLE

Submitted to the Office of Graduate and Professional Studies of

Texas A&M University
In partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Chair of Committee, Kausalai (Kay) Wijekumar

Co-Chair of Committee, R. Malatesha Joshi
Committee Members, Hersh Waxman
Julie Thompson
Head of Department, Michael de Miranda

August 2018

Major Subject: Curriculum and Instruction

Copyright 2018 Andrea Lynne Beerwinkle

 ABSTRACT

The ability to comprehend science texts is not only an academic skill but a life skill.

Currently, however, the majority of students across grade levels in the United States are

reading below grade level and have science achievement below grade level. The text

structure strategy, a reading comprehension strategy in which students are taught to use the

structure of a text (e.g., comparison, cause and effect, problem and solution) to construct the

main idea of the text, has been shown to be successful in improving reading comprehension.

Therefore, the text structure strategy was implemented in grade 7 science classes to improve

reading comprehension and science knowledge. The intervention included practice-based

professional development, weekly instructional planning, modeling of text structure lessons

in science, and adaptation of instructional materials to support the text structure strategy.

This study investigated the efficacy of the text structure strategy in a grade 7 science

classroom with four teachers and 169 students in a small, semi-rural district. The researcher

collaborated with school administrators to provide teachers with practice-based professional

development and ongoing in- and out-of-classroom support during the intervention. The

study utilized a quasi-experimental pretest-posttest design. Results from paired t-tests

showed that students significantly improved on science knowledge, reading comprehension,

signaling word knowledge, and main idea quality. The text structure strategy in science has

promise as an effective strategy to improve reading comprehension and science knowledge

of middle school students.

ii
 DEDICATION

This dissertation is dedicated to my person and my little people. There are not words

enough to express how much I love y’all. Thank you for supporting me on this journey.

This dissertation is also dedicated to my parents. My love of learning and asking

questions is solidly built on a foundation developed with your love and support. Thanks for

helping a stubborn opinionated girl become a determined knowledgeable woman.

iii
 ACKNOWLEDGMENTS

I want to acknowledge Dr. Kay Wijekumar for her unwavering support and

invaluable guidance over the past few years. It is a privilege and an honor to work with her.

iv
 CONTRIBUTORS AND FUNDING SOURCES

Contributors

This work was supervised by a dissertation committee consisting of Dr. Kausalai

(Kay) Wijekumar (advisor), Dr. R. Maltesha Joshi, and Dr. Hersh Waxman of the

Department of Teaching, Learning, and Culture, and Dr. Julie Thompson of the Department

of Educational Psychology.

The analyses depicted in Chapter IV were conducted in part by Shuai Zhang of the

Department of Teaching, Learning, and Culture.

All other work conducted for the dissertation was completed by the student

independently.

Funding Sources

This work was made possible in part by a grant from the Powell Foundation. Its

contents are solely the responsibility of the author and do not necessarily represent the

official views of the Powell Foundation.

v
 NOMENCLATURE

ELA English Language Arts

GSRT Gray Silent Reading Test

NAEP National Assessment of Educational Progress

SKA Science Knowledge Assessment

STAAR State of Texas Assessments of Academic Readiness

TEA Texas Education Agency

Text structure The organizational structure used by an author (e.g., comparison,

cause-and-effect, problem-and-solution, sequence, description)

vi
 TABLE OF CONTENTS

Page

ABSTRACT .......................................................................................................................ii

DEDICATION ................................................................................................................. iii

ACKNOWLEDGMENTS .................................................................................................iv

CONTRIBUTORS AND FUNDING SOURCES .............................................................. v

NOMENCLATURE ..........................................................................................................vi

LIST OF TABLES ............................................................................................................ix

CHAPTER I INTRODUCTION ........................................................................................ 1

Literacy Issues Affecting Adolescent Readers ................................................................ 2

Research Questions.......................................................................................................... 9

CHAPTER II LITERATURE REVIEW .......................................................................... 11

Review of Theoretical Framework ................................................................................ 11

Recent Content Area Literacy Interventions ................................................................. 15
Teacher Attitudes and Beliefs of Content Literacy ....................................................... 20
Science Literacy ............................................................................................................ 23

CHAPTER III METHODOLOGY ................................................................................... 24

Design ............................................................................................................................ 24
Participants .................................................................................................................... 24
Measures ........................................................................................................................ 27
Scoring ........................................................................................................................... 30
Procedures ..................................................................................................................... 32
Research Questions........................................................................................................ 39
Method ........................................................................................................................... 39

CHAPTER IV RESULTS ................................................................................................ 41

Research Questions........................................................................................................ 41
Descriptive Statistics ..................................................................................................... 41
Text Structure Intervention Effectiveness (Primary Research Question) ..................... 42
Effect of Gender, Ethnicity, Economic Status, Language Proficiency, and
Teacher on Intervention Effectiveness (Research Questions 2 and 3) .......................... 43

vii
 Effect of Science Knowledge Pretest on Learning Outcomes (Research
Question 4) .................................................................................................................... 48

CHAPTER V CONCLUSIONS ....................................................................................... 50

Research Finding in Context ......................................................................................... 50

Text Structure Intervention Effectiveness (Primary Research Question) ..................... 51
Effect of Gender, Ethnicity, Economic Status, Language Proficiency, and
Teacher on Intervention Effectiveness (Research Questions 2 and 3) .......................... 53
Effect of Science Knowledge Pretest on Learning Outcomes (Research
Question 4) .................................................................................................................... 55
Limitations of the study ................................................................................................. 55
Significance of the Study ............................................................................................... 56

REFERENCES ................................................................................................................. 58

APPENDIX ...................................................................................................................... 66

viii
 LIST OF TABLES

Page

Table 1 Chall’s Stages of Reading Development ................................................................ 5

Table 2 Themes Associated with Preservice Teachers’ Views on Literacy ..................... 22

Table 3 Descriptive Statistics of 7th Grade Student Body ................................................ 25

Table 4 Demographics of Students with Returned Consents at Wave 1 and Wave 2 ....... 26

Table 5 Descriptive Statistics of STAAR Grade 6 Reading Results for Students with
Returned Consents at Wave 1 and Wave 2 .......................................................... 26

Table 6 Researcher Designed Outcome Measures and Scoring Approaches with

Examples.............................................................................................................. 32

Table 7 Descriptive Statistics of Pretest and Posttest Scores ............................................ 42

Table 8 Paired t-Tests Comparing Pretest vs. Posttest Scores .......................................... 43

Table 9 Demographic Variable Effects on Signaling Word Posttests .............................. 44

Table 10 Demographic Variable Effects on GSRT Age-equivalent Posttests .................. 45

Table 11 Demographic Variable Effects on the GSRT Grade-equivalent Posttest ........... 45

Table 12 Demographic Variable Effects on the SKA Posttest Raw Score ....................... 46

Table 13 Demographic Variable Effects on Main Idea Competency Posttest Scores ...... 47

Table 14 Demographic Variable Effects on Main Idea Quality Posttest Scores............... 47

Table 15 Effect of Science Pretest Capacity on Gain Scores (Science Knowledge Raw
Scores) ................................................................................................................ 48

Table 16 Text Structure Strategy Use During Classroom Observations ........................... 49

ix
 CHAPTER I

INTRODUCTION

Middle school students are struggling with science reading comprehension. On

average one in four students have regularly failed the State of Texas Assessment of

Academic Readiness (STAAR) 8th Grade Science test since its beginning in the 2012-2013

school year (Texas Education Agency (TEA), 2018). Additionally, national science

achievement among grade 8 students is declining (National Assessment of Educational

Progress (NAEP), 2015). Students do not currently possess the reading comprehension

strategies needed to assist them now as they learn the science content presented in their

classrooms and later as they need to use the language of science to live healthy and

scientifically literate lives (i.e., the ability to understand scientific concepts needed for daily

life). Understanding the language of science is not only helpful in learning new concepts, it

is a critical part of science literacy (Lemke, 1990; Yore, Bisanz, & Hand, 2003). Norris and

Phillips (2003) found that reading and writing have a constitutive relationship with science,

meaning there is no science without reading and writing. O’Reilly and McNamara (2007)

further explored the connection between reading and science and found that students with

better reading comprehension were better able to compensate for low science knowledge

than students with low reading comprehension. Lemke (1990) supports the reading science

connection by suggesting that wording of science concepts may change from book to book

or teacher to teacher, but that “the pattern of relationships of meaning, always stays the

same” (p. x). Additionally, Yore et al. (2003) highlight that text structures, such as problem-

and-solution and cause-and-effect, are linguistic devices used in scientific texts. Therefore,

to help struggling students overcome their difficulty with science content area reading

1
comprehension, this study implemented a text structure-based intervention designed to

improve student science reading comprehension with a secondary aim of improving science

content knowledge.

Literacy Issues Affecting Adolescent Readers

Consistently low literacy levels. The majority of students in grade 12 (63%) and

grade 8 (66%) are reading at or below the basic reading level (NAEP, 2017). These numbers

are significantly worse for Black and Hispanic students and students with disabilities

(NAEP, 2017). According to The Nation’s Report Card (2017) students performing at or

below the basic level lack mastery of fundamental skills needed to perform at grade level.

This means the majority of adolescents in the United States have reading skills considered

below grade level and are likely to struggle with tasks such as making inferences, analyzing

what was read, summarizing a text, and other higher-level skills needed for success in school

and life. What is more troubling is since 1992 these numbers have remained relatively

unchanged. This means for over two decades the majority of adolescents in the United States

have been unable to read above a basic level and have entered the workforce or post-

secondary education with reading skills far below what is needed for success. Foorman,

Petscher, Stanley, Truckenmiller (2016) found similar struggles for students in grades 5 and

8. Conformity factor analysis of reading and language variables found 53% of fifth-grade

students and 72% of eighth-grade students scored low on all variables (Foorman et al., 2016).

Consistently low performance across multiple measures suggests that literacy practices at

the middle school level are not functionally meeting the needs of the majority of students.

Low science scores. In addition to low reading ability, students are performing at

critically low levels in science (NAEP, 2015). The majority of grade 8 students (66%) scored

2
at or the below basic achievement level (NAEP, 2015). Scores for Black and Hispanic

students are more troubling. Fewer than 20% of Black or Hispanic students achieved scores

at proficient or above. These numbers create a troubling picture of our nation’s current

knowledge of science.

Scores on high-stakes science tests are also critically low. In Texas, 26% of students

in grade 8 scored in the “did not meet grade level” performance standard category on the

STAAR grade 8 science test for the main administration in the 2016-2017 school year. This

means that one-in-four students were found to be “unlikely to succeed in the next grade or

course without significant, ongoing academic intervention” (TEA, 2017) according to the

STAAR performance label definition for “did not meet grade level”. Additionally, students

scoring in the “did not meet grade level” category “do not demonstrate a sufficient

understanding of the assessed knowledge and skills” (TEA, 2017). The number of students

scoring in the “did not meet grade level” category increases for Hispanic (31%) and African-

American (38%) students. Almost half (44%) of at-risk students and 70% of students

receiving special education services received a “did not meet grade level” score. Also

troubling is STAAR pass rates have remained stagnant since the test’s introduction in 2012

(TEA, 2018). Although the STAAR science test is presented as a measure of content

knowledge, all questions require grade-appropriate levels of reading comprehension for

students to understand the question being asked and to determine the correct answer.

Additionally, students must be able to determine whether the question is asking them to

perform tasks such as identify part of a cause-and-effect relationship or make a comparison

between given data.

3
 Issues in literacy instruction in middle school and the content areas. Literacy

instruction at the middle school level greatly differs from instruction at the elementary level

(Shanahan & Shanahan, 2008). In the middle grades, teaching is frequently

departmentalized. This means that students have a different teacher for each subject.

Additionally, middle school level content area teachers typically receive less pre-service

training in reading than early childhood through grade 6 (EC-6) or middle school English

Language Arts (ELA) teachers. In Texas, the reduced literacy training for content area

teachers may be due to the lack of literacy standards tied to grades 6-8 science courses.

Furthermore, mathematics is often suggested as the language of science through the focus

on formulas and calculations rather than oral or written language (Hand et al., 2003; Yore et

al. 2003). Despite lesson models such as the 5E model (engage, explore, explain, extend,

and evaluate) that encourage hands-on activities to teach science concepts, students gain

much of their information about science topics through reading (Kaldenberg, Watt, &

Therrien, 2015; O’Reilly & McNamara, 2007).

Following Chall’s (1983) stages of reading, see Table 1, middle school students

should be at the end of stage three and moving through to stage four. In stage three,

instruction focuses on introducing expository texts and strategies to extend reading

comprehension. In stage four, students increase their expertise in using strategies and deepen

knowledge in areas like text structures and features, genre, vocabulary, and other areas.

Students at the middle school level are thus expected to apply their well-honed reading

comprehension skills in not only their reading/ English Language Arts classrooms, but in

their science and social studies classrooms as well.

4
Table 1
Chall’s Stages of Reading Development
Stage Skills Taught Materials Used
Stage 0: Pre-reading Oral language development Picture books, alphabet books,
Birth to age 6 Letter names and sounds engage in pretend/pseudo-reading,
Phonological awareness writing, and language play

Stage 1: Initial reading/ Decoding Basic phonics skills Children’s storybooks, basal readers,
Grade 1 Six-syllable patters and trade books
Commonly used irregular words
Oral language development

Stage 2: confirmation and fluency Decoding Children’s storybooks, workbooks,

Grade 2-3 Fluency basal readers and trade books,
Affixes and roots familiar fiction and nonfiction
Analysis of multi-syllable words
Background knowledge development

Stage 3: Reading to learn Reading for meaning Children’s literature, basal readers,
Grades 4-8 Expository text structures workbooks, content-area textbooks,
Reading strategies beginning reference materials, and
Background knowledge development Internet sources.

Stage 4: Reading for meaning Fiction and nonfiction, reference

Multiple viewpoints Inferential thinking materials, newspapers, magazines,
Grades 9-12 Perspective and Internet sources
Specialized vocabulary

Stage 5: Verbal reasoning Fiction and nonfiction, periodicals,

Construction and reconstruction Inferential thinking journals, and Internet sources.
College Author’s perspective
Analysis of genres
Adapted from Farrall, M.L. (2012)

Middle school students are expected to be solidly in the reading to learn phase,

having mastered the learning to read phase in lower grades. Yet, based on NAEP (2017) data

there is a problem with this expectation. The majority of middle school students are

struggling with their reading ability and need help (NAEP, 2017). Frequently, assistance for

these students is offered through an elective reading class or a regular ELA class

purposefully populated with struggling readers (Swanson, Wanzek, Vaughn, Roberts, &

Fall, 2015). However, the majority of students with learning disabilities spend 80% of their

time in the general education classroom, most frequently science and social studies class

(Aud et al., 2010). A drawback of struggling students spending time in general education

5
content area classes is the limited help offered in these classes with regard to improving

reading comprehension in these areas. This is a problem because as Kaldenberg et al. (2015)

point out, despite a shift towards hands-on learning, students are still expected to gain most

science knowledge through expository texts.

Compounding the problem for content area teachers and struggling students is the

variety of reading struggles students face (Swanson et al., 2015). Students may struggle with

decoding, vocabulary, fluency, and background knowledge deficits, all of which affect

reading comprehension (Swanson et al., 2015). Gersten, Fuchs, Williams, and Baker (2001)

suggest that students with learning disabilities are able to identify when they do not

understand what they read, but they are not able to use strategies effectively to fix the

problem. Students need specific reading strategy instruction to target their difficulties so that

students can automatize these strategies and have them available when texts become difficult

(Cantrell, Almasi, Carter, Rintamaa, & Madded, 2010). However, students being able to use

multiple strategies is not the end goal. The end goal is that students become strategic readers.

Strategic readers think about the reading context and are efficiently and effectively able to

apply the best strategies based on the context.

The need for changes in content literacy instruction. Reading comprehension of

middle school science texts is especially critical. The topics covered in grades 6-8 science

courses provide students with basic and functional science knowledge necessary to have a

healthy life (Berkman et al., 2004; TEA, 2011), understand their environment (TEA, 2011),

and be good stewards of the earth (TEA, 2011). It is therefore imperative that students be

able to comprehend and later recall the information they gain in these courses not only to

pass high stakes tests but to live scientifically literate lives.

6
 Additionally, several researchers (Anderson, 1999; Hand et al. 2003; Lemke, 1990;

Norris & Phillips, 2003, Yore et al., 2003) argue that text is actually the system most used

by scientists to complete science tasks and that science literacy requires being able to read

and write in the language of science. Lemke (1990) argues that the language used in science

is a result of the members of the science community agreeing on a set way of communicating

about science. However, students are not yet fluent in the communication processes and

patterns used in science. This lack of fluency in the language of science means that students

are in critical need of reading comprehension strategies that will help them develop both

fundamental and derived science literacies needed for the immediate need of learning

science concepts in the classroom and the lifelong need of being able to converse

knowledgeably about science topics.

As stated previously, many students struggle with decoding and vocabulary deficits

even at the middle school level. Poor decoding skills place additional strain on working

memory needed for constructing meaning and monitoring reading comprehension (Baker,

DeWyngaert, & Zeliger-Kandasamy, 2015). Nevertheless, fluent reading alone is not enough

for full reading comprehension. It is common for a teacher to be surprised that a student,

who appears to be a good reader, because they can decode grade-level words, does not

comprehend what is read (Cartwright, 2015). Research has shown that better readers have

better reading comprehension monitoring or metacognitive awareness than weaker readers

and that these skills do not automatically improve as students get older and read more (Baker

et al., 2015). Decoding and vocabulary interventions alone are not enough to help students

improve reading comprehension.

7
 In fact, over a twenty years ago, Pressley and Wharton-McDonald (1997) expressed

concerns about attention being paid to word-level processing at the cost of text

comprehension. Students need specific reading comprehension/metacognition instruction to

improve reading comprehension. As mentioned previously, content area teachers may not

possess the knowledge to provide this type of instruction. Authentic texts are rarely, if ever,

transparent about the structure of ideas. Students must make inferences about how words,

sentences, paragraphs, and longer chunks of texts are connected and how that relates to the

main idea (Baker et al., 2015). Further, Pressley and Wharton-McDonald (1997) argue that

good readers deliberately look for connections between different parts of a text and attempt

to integrate those into a whole, sometimes using the connections to make inferences that fill

in gaps in the text. However, struggling students do not develop strategies that help them

make these connections without instruction (Pressley & Wharton-McDonald, 1997). The

text structure strategy can help students improve in making these connections and inferences

by providing explicit instruction to improve reading comprehension.

Reading is a critical lifelong skill, not simply a tool used to pass tests. The ability to

comprehend what is read is needed to correctly take medicine, compare insurance policies,

evaluate political candidates, and a host of other day-to-day tasks. Reading comprehension

is then not only a literacy skill but also a life skill. Content areas, such as science, provide

students with information and skills critical for full participation in adult life. The sharing of

scientific knowledge is a social practice done through text and reading (Hand et al. 2003;

Lemke, 1990). The dependence of science on sharing ideas via text also means that

individuals with poor reading comprehension are also limited in their ability to gain science

knowledge (Norris & Phillips, 2003). Without the ability to comprehend scientific texts and

8
then interpret and evaluate the claims within them, students will be limited in their ability to

learn how to best care for an ailing pet, increase the yield of their tomato plants, or limit the

potential healthcare costs of a society that eats more and moves less. To ensure that

adolescents are leaving the K-12 environment fully armed with the skills needed to be

successful citizens, educators must reevaluate how reading comprehension is addressed in

the middle grades and the content areas.

To address the previously discussed issues, this dissertation study used the text

structure strategy to provide an intervention designed to improve middle-grade students’

reading comprehension and recall of science content-area text. The theoretical foundation

for this study comes from the text structure work of Bonnie Meyer (Meyer, 1975) which has

seen positive results with students in grades 4-7 (Meyer et al., 2002; 2010; Meyer,

Wijekumar, & Lin, 2011; Wijekumar, Meyer, & Lei, 2017; Wijekumar et al., 2017).

Although these studies have used expository texts, they have done so in the English

Language Arts classroom. This study sought to improve reading comprehension of science

texts by focusing on reading instruction in the science classroom.

Research Questions

This dissertation study addresses three factors in student reading comprehension of

science texts: 1) knowledge of text structures and signaling words, 2) competency in using

signaling words and text structures to aid in reading comprehension, and 3) disciplinary (i.e.,

science) literacy.

This study aimed to evaluate the effectiveness of a text structure strategy-based

intervention within the context of middle school science courses. The study sought to answer

one main question:

9
1. Were there pre-posttest improvements in science and reading comprehension scores for

seventh-grade students using the text structure strategy for 10 weeks?

This study also sought to answer three secondary questions:

2. Were there differences in improvements based on gender, ethnicity, economic status,

and/or language proficiency?

3. Were there any teacher effects on student learning outcomes (e.g., Did student

performance differ by science teacher?)

4. Do students with lower pretest science scores (i.e., raw score below 12 or 41%)

demonstrate better improvements after intervention than those with higher pretest

scores?

10
 CHAPTER II

LITERATURE REVIEW

In the introduction, I outlined the problems facing students in regards to reading

comprehension and content literacy with science. In this section, I will review the literature

relating to the theoretical framework for the study, present the literature relating to content

area literacy interventions for science, synthesize the literature on teacher influence of

content literacy practices, and identify the gaps in the current literature, see Table A-1 in the

Appendix for research design, population, measures, duration and conditions, and results of

included studies.

Review of Theoretical Framework

The text structure strategy. The theoretical framework for this study has its roots

in what Meyer, Brandt, and Bluth (1980) refer to as the text structure model. The text

structure strategy stems from research showing that information presented higher in the

content structure of a text is connected to better recall than information presented lower in

the content structure (Meyer, 1975). For example, a science textbook passage comparing

three of Earth’s biomes in terms of climate, animals, and plants the information higher in the

content structure would be the comparison of climate, animals, and plants. Information lower

in the context structure would be the causes and effects of how the plants or animals have

adapted to the climate.

The text structure strategy is the use of the content structure of a text to organize

information in the memory, which in return helps with recall (Meyer et al., 1980).

Additionally, the use of the text structure strategy without prior instruction has been found

to be used by those with good reading comprehension, while those with poor reading

11
comprehension tend to engage in listing or knowledge telling (Meyer et al., 1980). Good

comprehenders are more likely to use the same top-level structure as the text when recalling

what was read than poor comprehenders (Meyer et al., 1980). For example, good

comprehenders are more likely to use the comparison structure of the biomes passages from

the previous example when recalling the passage than poor comprehenders. Furthermore,

following the top-level structure of a passage can provide students with “a systematic

learning and retrieval guide” (Meyer et al., 1980, p.99). Readers can use the comparison

structure of the biomes passage to help them better mentally organize and retrieve

information about the biomes being compared. The text structure strategy has been heavily

studied (Meyer et al., 2002, 2010; Meyer & Wijekumar, 2014; Wijekumar et al., 2014;

Wijekumar, Meyer, & Lei, 2012, 2013, 2017) and found to improve structure strategy

competency and reading comprehension. However, the structure strategy has not been

thoroughly explored within the science classroom as a way to improve both science literacy

and science achievement. The study worked with seventh-grade science teachers on

integrating the text structure strategy into daily instruction. The focus on teacher-led text

structure strategy instruction over intelligent-tutor led instruction is also a change from

recent studies with the text structure strategy and adds to the literature on the text structure

strategy as a teacher-led intervention.

Reading skill and science achievement. Another element of the theoretical

framework of this study is the potential for reading skill to compensate for low content

knowledge. In a study on the impact of science knowledge, reading skill, and reading

strategy knowledge, on the science achievement of over 1,600 high school students O’Reilly

and McNamara (2007) found that reading skill was significantly correlated to multiple

12
measures of science achievement and that the effect size of reading skill on science

achievement measure was moderately large.

O’Reilly and McNamara (2007) measured science knowledge with a researcher-

developed 18-item multiple-choice test. Cronbach’s alpha for the measure was α= .74.

Reading skill was measured using the Gates-MacGinitie reading skill test level 7/9.

Cronbach’s alpha for the Gate-MacGinitie in the study was α= .95. Reading strategy

knowledge was measured with a 25-item multiple-choice questionnaire. Cronbach’s alpha

was α= .72. Science achievement was measured with 8 multiple choice and 12 open-ended

questions over a science passage about meteorology. Cronbach’s alpha for the multiple-

choice questions was α= .83 and α= .87 for the open-ended questions. Students’ course grade

and achievement on the Virginia standardized science tests were also used to measure

science achievement.

O’Reilly and McNamara found that reading skill had moderate positive correlations

with the multiple-choice questions, r(1,442)=.527, and open-ended questions, r(1,345)=

.641, measures of science achievement and a moderate correlation with a state standardized

science text measure of achievement, r(692)=.582. These findings support that improvement

in reading comprehension can potentially help improve science achievement. Reading was

also moderately correlated with science knowledge, r(1433)=.577. This finding is relatively

unsurprising since reading is the main way students acquire new science knowledge, and it

is difficult to acquire knowledge with poor reading comprehension.

O’Reilly and McNamara also ran regressions for each measure of science

achievement. Reading skill was a significant predictor for the multiple-choice question (β =

.379), open-ended question (β = .499), and state science test models (β = .395). For each of

13
the previously mentioned models, reading skill contributed the most to the model. Again,

this shows the positive possibilities of a reading comprehension intervention on improving

science achievement on multiple fronts. The effect sizes of reading skill on the multiple-

choice question (d= .64), open-ended question (d= .78), and state science test (d= .73)

measures were moderately large. The effect size of reading skill on science achievement for

students with low science knowledge was moderate to large (multiple-choice question, d=

.46; open-ended question, d= .89; state science test, d=.96). O’Reilly and McNamara also

found that skilled readers with lower science knowledge scored higher on the science

achievement test than students who were less skilled readers yet had higher science

knowledge.

This study provides empirical support for the impact of reading skill on science

achievement and, more specifically, that reading skill can compensate for lower science

knowledge. Further, this study is included in the theoretical framework of the current study

because O’Reilly and McNamara show the impact of reading skill on science achievement.

O’Reilly and McNamara’s findings suggest that skilled readers with low science knowledge

may be better able to leverage their reading skills in making inferences that help compensate

for a lack of background knowledge, whereas less skilled readers students, despite high

science knowledge are less able to harness their science knowledge to compensate for

decreased reading comprehension. This study also adds support to the use of the text

structure strategy to improve science achievement.

Studies on content area reading have shown success within and outside of content

area classrooms. Vaughn, Martinez, and Wanzek (2017) found small to moderate effect sizes

for social studies content knowledge acquisition (ES=.40) and social studies content-related

14
reading comprehension (ES=.20). Vaughn et al.’s study supports the promise of a reading

comprehension strategy in a content area class. Additionally, Wijekumar, Meyer, and Lei

(2017) found positive outcomes in reading for middle school students. Although Wijekumar

et al. (2017)’s intervention took place during the English Language Arts period, the texts

used in the intervention were content area texts and the researcher designed measure used

science texts. Wijekumar et al. show strategy instruction can have a positive impact on

content area reading comprehension as measured by standardized and researcher designed

measures. The text structure strategy can help support reading comprehension by providing

students with an explicit structure for mentally organizing texts, which in turn, results in

improved meaning-making (Meyer & Ray, 2011). Based on the empirical evidence, the text

structure strategy seems to be a promising approach in making significant impacts on science

literacy and science achievement.

Recent Content Area Literacy Interventions

Science interventions. Seifert and Espin (2012) explored the effects of a text-

reading, vocabulary learning, combined (text-reading and vocabulary learning), or control

condition on primary outcomes of reading fluency and vocabulary knowledge and a

secondary outcome of reading comprehension. The main goal of the study was to identify

interventions that had a direct effect on the ease of reading scientific texts by improving

fluency and vocabulary for students with learning disabilities. A secondary goal was to

improve reading comprehension of scientific texts. Participants in the study were 20 students

with documented learning disabilities in grade 10 enrolled in regular education biology

classes. Treatment sessions lasted 45 minutes (30 minutes for intervention and 15 minutes

for assessment) and the control session lasted 15 minutes (assessment only) and took place

15
over four days with two sessions a week for two weeks. Students in the text-reading

condition (designed to improve fluency) and the combined text-reading and vocabulary

condition had significantly higher scores on word reading than students in the vocabulary

only or the control conditions. Similarly, the vocabulary and combined conditions had

significantly higher scores on the vocabulary measure than the text-reading or control

conditions. There was no significant difference for reading comprehension.

During the intervention, Seifert and Espin (2012) used passages from a standard

biology textbook that was used in four of the five schools that participating students attended,

and the study shows promise for using science texts as part of focused interventions aimed

at improving science-reading skills for students with learning disabilities. However, the

information was not explicitly connected to classroom instruction. Whereas, the text

structure intervention used by the current study builds on the use of science passages by

utilizing the intervention with passages from current areas of classroom instructional focus

in the current science textbook adopted by the school. Seifert and Espin (2012) also found

significant results for students with learning disabilities. The text structure intervention

extends on Seifert and Espin’s success with students with learning disabilities and teaches

the text structure intervention to all students, because current NAEP results show that the

majority of students in grades 4, 8, and 12 are struggling with reading comprehension

regardless of disability diagnosis (NAEP, 2017). Additionally, Seifert and Espin’s study

focused on reading fluency and vocabulary knowledge with a secondary focus of reading

comprehension, as a result, Seifert and Espin did not include explicit instruction in reading

comprehension strategies or identification of the main idea. In contrast, the main focus of

the text structure intervention is improved reading comprehension. This focus is achieved

16
by teaching students to use the structure of a text to scaffold reading comprehension and

assist in finding the main idea. Finally, Seifert and Espin’s intervention was implemented by

two graduate students and Ms. Seifert outside of classroom instruction; classroom teachers

were not involved. The text structure intervention, however, is implemented by participating

classroom teachers and during classroom instruction in an effort to promote the sustainability

of the intervention once the study has ended. Seifert and Espin’s study supports investigation

into more specific treatments to improve reading comprehension of science texts for students

with learning disabilities and as well as students without learning disabilities. This is also

supported by the consistently low percentage of students receiving special education services

that pass the STAAR grade 8 science test (TEA, 2017). While it is important that students

be able to read science texts with ease, if students are not able to comprehend what they read

then reading ease is relatively pointless. The text structure strategy fits nicely in this gap

because it both provides a specific treatment focused on reading comprehension that can be

applied to texts of any topic, including science, and is relatively simple in nature allowing

the intervention to be implemented by varying instructors.

Cervetti, Barber, Dorph, Pearson, and Goldshmidt (2012) compared the effects of an

integrated approach to science and literacy to separate science and literacy instruction on

grade 4 students’ science understanding, reading comprehension, science vocabulary, and

science writing. The teachers in the study were 94 fourth grade teachers that were randomly

assigned to either treatment or comparison conditions. The treatment condition included four

investigations that occurred over 10 sessions each, totaling 40 sessions over 22 weeks.

Sessions were 45 to 60 minutes in length. Students in the treatment condition significantly

17
outperformed the separate science and literacy condition in science understanding, science

vocabulary, and several aspects of science writing.

Teachers in Cervetti et al.’s (2012) intervention were provided with a researcher-

developed unit on light. In contrast, the text structure intervention focuses on teaching both

teachers and students to use the text structure strategy with any text, thus allowing the

intervention to be sustainable when researchers are not present. During the reading and

writing activities in Cervetti et al.’s intervention, students were provided with “explicit

instruction in the use of targeted inquiry skills and literacy strategies” (p. 639) and were

taught to identify keywords in texts and then use them to build the main idea. The text

structure intervention is different from Cervetti et al.’s in that the text structure intervention

explicitly teaches one reading comprehension strategy that subsumes comprehension skills

such as making predictions, summarization, and inferencing. Additionally, the text structure

intervention uses the macro-structure (e.g., comparison, cause-and-effect, problem-and-

solution) of the passage to guide students in finding keywords (signaling words) to help in

constructing the main idea.

Fang and Wei (2010) explored the efficacy of an inquiry-based science curriculum

with reading (ISR) in developing students’ fundamental and derived science literacies

compared to an inquiry-based science (IS) curriculum alone. Participants were 233 grade 6

students nested in two science classrooms. Both the ISR and IS conditions participated in an

inquiry-based science curriculum for 50 minutes per day for 22 weeks. The ISR condition

also included 15-20 minutes of reading strategy instruction on Thursdays and a home science

reading program (HSRP) in which students selected one high-quality science book to take

18
home for the week and share with a family member. Students in the ISR condition

significantly outperformed students in the IS condition on all measures.

Fang and Wei’s work shows the positive effect that literacy instruction in science

can have on overall reading ability and science knowledge. Fang and Wei also point out that

they worked closely with the science teachers. The text structure intervention follows this

model of working closely with science teachers to implement reading strategies to improve

student science reading outcomes. However, the strategies used in Fang and Wei’s

intervention changed week-to-week depending on what the researchers and science teachers

felt was the best strategy for the current topic. As a result, it is difficult to determine which

strategies had the greatest impact on student progress. The text structure intervention, on the

other hand, focuses on a single reading comprehension strategy. Focusing on a single

strategy is important to developing and promoting only the strategies that have proven

successful in and of themselves.

Combined science and social studies interventions. Boardman, Klingner, Buckley,

Annamma, and Lasser (2015) studied the effects of using Collaborative Strategic Reading

(CSR) in middle school science and social studies classrooms on students (n=1074) in 19

classrooms in a large urban district during the whole school year. Due to scheduling

conflicts, students within the CSR condition were further split into full and partial CSR.

Students in the full CSR condition participated in a CSR lesson one day each week in both

their science and social studies classes, and students in the partial CSR condition received

only one CSR lesson each week in either science or social studies. Small effect sizes were

found for both the full and partial CSR conditions on all measures.

19
 Boardman et al. (2015)’s Collaborative Strategic Reading utilizes several strategies

throughout the reading process. Because multiple strategies were used, students were also

required to remember multiple strategies, and it is difficult to know the individual effects of

each strategy. As previously mentioned, the text structure intervention teaches students to

use one strategy that subsumes reading comprehension skills such as getting the gist,

summarizing, and making inferences. Boardman et al. (2015) taught student to get the gist

of the text by identifying the most important who or what in the section and important

information associated with the who or what. While this strategy is more explicit than the

main idea strategy often given by textbooks, it is still somewhat vague. The text structure

intervention provides students with explicit guidance on stating the gist by teaching students

to use the structure of the text to find key information and then use a text structure specific

stem to write the main idea (i.e., The cause was _____ and the effect was _____). A key

feature of CSR is collaborative reading where each student in a group is responsible for a

strategy. This takes aspects of reading comprehension out of the reader’s hands. In contrast,

students are taught to use the text structure strategy on an individual level which mirrors

how students will be expected to read and learn most of their lives. While CSR shows

promise for improvement of content-area knowledge and reading comprehension, the

transfer of skills to content areas needs more attention to improve the learning outcomes.

Teacher Attitudes and Beliefs of Content Literacy

Content-area interventions have shown promise for improving comprehension,

vocabulary, and content knowledge (Boardman et al, 2015; Cervetti et al., 2012; Fang &

Wei, 2010; Seifert & Espin, 2012). However, content-area instruction does not take place in

20
a vacuum. Teacher beliefs and skills can and do influence the instruction students receive

(Kukner & Orr, 2015; Yore, 1991).

In a study of 16 pre-service content areas teachers, Kukner and Orr (2015) found

several themes that differentiated pre-service teachers (PSTs) committed to literacy in the

content-area and those less certain about the role of literacy in their classroom. These themes

are presented in Table 2. Unfortunately, students will face both teachers committed to

literacy as well as those less certain about literacy. However, the structure strategy provides

benefits to both teacher types. Teachers less certain about literacy in the content-area

classroom are provided with clear concrete methods of infusing the strategy into everyday

lessons and ways to model and practice the strategy are shown and modeled during

professional development sessions. Further, the structure strategy may be more appealing

than other strategies to content-area teachers uncertain of using literacy strategies in their

classroom. The structure strategy focuses on organizing the content to make the connection

between ideas clearer which is likely to seem valuable to teachers less comfortable with

literacy practices. The structure strategy also supports teachers committed to using literacy

practices in their classrooms. A key theme from these teachers was that they saw the

connection between literacy routines and thinking and learning. The structure strategy

encourages students to think about the ways ideas are connected in a text and provides

students with a means of enhancing learning by using those connections to make better

inferences about science ideas and topics.

21
Table 2
Themes Associated with Preservice Teachers’ Views on Literacy
Strong commitment to literacy
Expanded understandings of literacies
Literacy routines as opportunities for thinking
and learning
Clear connections to curriculum outcomes and
relevant authentic assessments
Less certain about role of literacy in teaching
Inability to speak fluently about incorporating
literacy into teaching
Lack of metacognitive awareness on how to plan
for literacy strategies
Lack of awareness of need to model literacy
strategies

Yore (1991) conducted a survey of secondary science teachers’ attitudes and beliefs

about reading instruction in science. The Science and Reading Questionnaire was completed

by 215 grade 6-12 science teachers in British Columbia. The questionnaire had 25 7-point

Likert items with a 12-item attitude subscale and a 13-item belief subscale. On the attitude

subscale, respondents generally agreed that science teachers have a responsibility to teach

reading in the science content. The majority of respondents agreed that teachers should help

students improve reading, be familiar with the theories of the reading process, teach students

how to read science materials, and help students move past the literal and toward the

interpretive level. However, the majority of teachers did not feel that certification for science

should include knowledge of reading instruction. On the belief subscale, respondents

generally agreed that reading is more than decoding words on a page. Teachers generally

agreed that science background knowledge was needed to move science reading beyond

memorization and that students used familiar examples when trying to comprehend new

concepts. Similarly, the majority of teachers rejected the idea that a science text required no

background knowledge for reading comprehension. A near majority of teachers also agreed

that poor readers did not follow the logical structure of paragraphs. This study shows promise

for teacher uptake of a reading comprehension intervention in the science content. Science

22
teachers appear to be open to the idea of helping students improve their science literacy and

see doing so as a responsibility of the job. Additionally, some teachers also seem to

understand the connection between understanding text organization (logical structure of

paragraph) and reading comprehension ability.

Science Literacy

Norris and Phillips (2003) argue that reading and writing are “inextricably linked to

the very nature and fabric of science, and, by extension, to learning science” (p.226). They

also make the case that there are two types of science literacy: fundamental and derived, and

that fundamental science literacy is often ignored, which is the thesis of their paper.

Fundamental science literacy is defined as being able to read and write in the science content.

Derived science literacy is defined as being knowledgeable about science content. In a

review of science literacy definitions, Norris and Phillips found that the majority dealt with

derived science literacy and that few definitions focused on the ability to read or write in

science. Norris and Phillips suggest that a cause for the lack of attention to fundamental

science literacy is science teachers’ view of reading as a simple process and text as

transparent because they are not struggling readers nor is their main focus the study of

reading. In further defining fundamental science literacy, Norris and Phillips argue against

a view of reading based solely on decoding and finding information. Instead, they favor a

view of reading that includes reader background knowledge, decisions on what is relevant

in the text, and requires the reader to create meanings and inferences and contextualize

information. The text structure strategy supports this view of reading by assisting students

as they decide what is relevant in the text and then as they make inferences supported by that

information.

23
 CHAPTER III

METHODOLOGY

Design

This quasi-experimental study with pretest and posttest investigated the effects of a

text-structure intervention on reading and science achievement of grade 7 students in a rural

middle school. The demographics of the school as well as the grade 7 student body are

presented in Table 3. The science teachers at the middle school volunteered to implement

the text structure intervention during their daily science instruction for 10 weeks starting in

November and ending in February. The middle school science classes were organized with

three teachers teaching six sections each and one teacher teaching four sections. A team from

a large research university in the southwest United States recruited the middle school to

participate in the study through a university community relations coordinator. Participating

teachers consented to participate and were required to complete two days of professional

development related to the study.

Participants

The students. All students enrolled in a grade 7 science class were invited to

participate in the study. Parental consent forms were sent home in two waves via the science

teachers. In the first wave, 89 students returned consent forms for the pretest. After the

second wave, a total of 173 students had returned the signed consent documents to participate

in the study for posttest. There were no statistically significant differences between students

who submitted consent in wave 1 and wave 2 on demographic variable. Student

demographics are presented in Table 4. The STAAR grade 6 reading test scores, see Table

5, indicate that 60% of students with available scores (n=49) at wave 1 reached the

24
approaches level (the minimum level considered passing) and had an average percentile of

47.30. At wave 2, 56% (n=86) of students with available scores had reached the approaches

level. The reported Lexile based on STAAR grade 6 reading test scores for students at wave

1 and wave 2 ranged from 445L to 1500L. The average Lexile for students at wave 1 was

925L and the average level at wave 2 was 895L. Although there is no direct relationship

between Lexile and grade level, 855-1165L represents the range for the middle 50% of

readers measured in the middle of grade 6 (Lexile, 2018). Therefore, students before the start

of the intervention were reading far below, at, and above grade level. At pretest, 10 were in

teacher A’s class, 24 were in teacher B’s class, 40 were in teacher C’s class, and 15 students

were in teacher D’s class. At posttest, 13 were in teacher A’s class, 56 were in teacher B’s

class, 72 were in teacher C’s class, and 28 students were in teacher D’s class.

Table 3
Descriptive Statistics of 7th Grade Student Body
n %
Total School Enrollment 895
Grade 7 Enrollment 456
Female 215 47.15
Race/Ethnicity
African American 113 24.78
Caucasian 165 36.18
Hispanic 162 35.53
Asian 5 1.1
Two or more races 11 2.41
Economic disadvantage 358 78.29
Special education 41 8.99
Other health impairment 7 1.54
Learning disabilities 18 3.95
Autism 7 1.54
Emotional disturbance 3 0.66
At-risk 274 60.09

25
Table 4
Demographics of Students with Returned Consents at Wave 1 and Wave 2
Wave 1 Wave 2
n % n %
Female 47 52.81 102 58.96
Race/Ethnicity
African American 17 19.10 44 25.43
Caucasian 40 44.94 71 41.04
Hispanic 30 33.71 50 28.90
Asian 1 1.12 4 2.31
Two or more races 1 1.12 4 2.31
Economic disadvantage 61 68.54 126 72.83
Limited English Proficiency 11 12.36 16 9.25
Special Education
Other health impairment 0 0 2 1.16
Learning disability 1 2.25 2 1.16
Autism 1 1.12 2 1.16
Emotional disturbance 1 1.12 1 0.58
At-risk 41 46.07 94 54.34
Teacher
A 6 6.74 14 8.09
B 32 35.96 54 31.21
C 36 40.54 76 43.93
D 15 16.85 29 16.76

Table 5
Descriptive Statistics of STAAR Grade 6 Reading Results for Students with Returned
Consents at Wave 1 and Wave 2
Wave 1 Wave 2
n M SD n M SD
Raw score 79 25.35 8.10 158 24.78 8.04
Percent score 79 62.65 20.27 158 62.11 20.26
Approaches 44 103
Meets 26 59
Masters 17 31
Percentile 79 46.13 29.60 158 44.05 28.52
Lexile 79 909.56 231.99 158 891.15 236.41

The teachers. Teachers completed their consent forms during the first professional

development. The four science teachers comprised the entire grade 7 science department.

26
The teachers agreed to participate in the study by using text structures in their science lessons

as appropriate. All teachers were women. One teacher was African American, two teachers

were Caucasian, and one teacher did not know her race or ethnic background. Teacher A

was new to teaching and was hired in November, prior to the start of the intervention. Prior

to her hire, her students had been instructed by a long-term substitute. She held two Bachelor

of Science degrees- one in biology and one in cytogenetics. She also held an MBA. Teacher

C had 30 years of teaching experience, all teaching science. She held a bachelor’s degree in

secondary education with a science composite certificate. Teacher D had four years of

teaching experience with three of those years spent teaching science. She held both a

bachelor’s degree and a master’s degree in exercise science. Teacher B had five years of

teaching experience with four teaching science. She held a bachelor’s degree in

interdisciplinary studies with an emphasis in 4-8 math and science.

Measures

Demographic variables. The students’ special education indicator, economic

disadvantage status, age, LEP status, gender, race/ethnicity, classroom, and class period

were collected.

Standardized tests of reading comprehension. Reading comprehension was

measured with the Gray Silent Reading Test (GSRT) (Wiederhold & Blalock, 2000) at

pretest and posttest. This test was selected because it can be used in a group setting, has two

forms, and tests comprehension processes, including finding the main idea. The GSRT is a

multiple choice reading comprehension test comprised of 13 short passages that increase in

complexity followed by five multiple-choice questions. This test is designed for readers age

seven through 25. Alternate-form reliability was reported in the test manual at 0.85.

27
Cronbach’s α for form A and B were 0.95 and 0.94, respectively. All five major text

structures are used in the GSRT in a minimum of one passage. Additionally, the multiple-

choice questions necessitate a rephrasing of the information presented in the passage. In this

study, Form B was administered at pretest and Form A was administered at posttest. The

pretest GSRT score was used as a covariate for data analyses when examining the effects of

the text structure intervention on the dependent measures of reading comprehension. The

GSRT posttest score was the outcome measure for the second research question.

The STAAR grade 6 reading test was also used as a standardized test of reading

comprehension. The measure is administered state-wide and is reported to be valid and

reliable. Prior released tests show high reliability (Cronbach’s alpha= 0.91). The test

includes narrative and expository passages with multiple-choice questions measuring

vocabulary and reading comprehension for each grade level. The STAAR grade 6 test was

also used as a covariate for data analysis when examining the effects of the text structure

intervention on the dependent measures of reading comprehension.

Standardized test of science knowledge. To measure students’ current science

knowledge, a researcher created science knowledge assessment (SKA) using released

STAAR grade 8 science test questions was administered at pretest and posttest. Using the

Texas Essential Knowledge and Skills (TEKS) as a guide, all released STAAR grade 8

science tests were reviewed for questions that were listed by the Texas Education

Association (TEA) as addressing grade 7 and 6 TEKS. Only questions addressing grades 6

and 7 TEKS were used because students could reasonably be expected to have learned

information tested in questions covering grade 6 TEKS the previous year and to learn

information tested in questions covering grade 7 TEKS during the course of the current

28
school year. Students could not reasonably be expected to know the material in questions

covering grade 8 TEKS. Questions from the released tests that assessed grade 6 and 7 TEKS

were then compiled. The resulting list of test questions was then divided into two sets of test

questions. With exception of TEKS 7.6(B) and 7.11(A) which were only tested once across

all released tests, all TEKS were tested a minimum of two times. Table A-2 in the Appendix

lists the TEKS assessed in the modified test. Validity and reliability of the STAAR tests

were established by an independent company contracted by TEA. The reported Cronbach’s

alpha reliability of the STAAR grade 8 science test is 0.91 and an expert review panel agreed

that test items were 97.7% aligned with expectations.

Experimenter-designed measures of reading comprehension. Student use of

problem and solution and comparison text structures was tested using two equivalent test

forms. One form was administered before teachers began using the text structure intervention

in their classrooms and the second form was administered immediately after the conclusion

of the intervention. Each form had two passages: one passage using the problem and solution

text structure and one passage using the comparison text structure. Identification of the top-

level structure and competence were assessed with both the problem and solution and

comparison passages. Signaling word identification was measured using the comparison

passage. Additionally, the comparison passage had the additional variable of number of

issues compared.

Problem and solution text structure passage. The two problem and solution

passages dealt with a) rats and b) dogs. The passages were equivalent; each having 98 words,

72 idea units, and equivalent scores on readability, text structure, and signaling (Meyer,

2003). Each problem and solution passage presented students with a problem that was

29
relatively unfamiliar, the cause of the problem, and a solution that eliminated the cause of

the problem. Students were asked to read the passage, remove the passage from their sight

by putting it in an envelope, and then write all that they could recall about the problem and

solution passage. The dependent variables for the problem and solution passage are top-level

structure and competency in using the problem and solution text structure to organize the

recall.

Comparison passage. The topic of the comparison passage on the pretest or posttest

was either a) Hagar Qim Stone Circles compared to Stonehenge or b) Mt. Rushmore

compared to Easter Island. Students were asked to complete three tasks using the comparison

passage: 1) a fill in the blank cloze task with 5 blanks testing knowledge of comparison

signaling words, 2) a main idea task where students were asked to write a main idea no

longer than two sentences with the passage in view, and 3) a recall task with the passage out

of view, similar to the problem and solution recall task. Dependent variables for the

comparison passage included top-level structure, comparison competency, number of issues

compared, and signaling word test scores.

Text structure strategy use in classrooms. Use of the text structure strategy was

recorded during each classroom visit. During each visit, if the text structure strategy was

appropriate for the lesson, if the teacher used the text structure strategy during the lesson, if

I used the text structure strategy during the lesson via co-teaching, and if I modeled the

lesson (i.e., I taught the whole lesson).

Scoring

GSRT. The GSRT was scored using the scoring procedures stated in the testing

manual.

30
 Science knowledge assessment. The SKA was scored using the answers provided

by TEA for each released STAAR grade 8 science test.

Experimenter-designed measure of reading comprehension. The signaling word

responses for the comparison passage were scored using computer algorithms. Correct

answers for each blank were given a score of 7 with a maximum possible score of 35.

The top-level structure, competence, quality, and number of issues compared

measures were scored by a trained member of the research team. Problem and solution and

comparison competence from the main idea and two recall tasks were scored using manuals

from previous text structure intervention studies (Meyer et al., 2010; Wijekumar et al.,

2014). The recalls were scored on an 8-point scale and the main idea was scored on a 6-point

scale. Both recalls (problem and solution, comparison) and the main idea (comparison)

assessed students’ proficiencies in using the text structures as taught during the text structure

intervention. Table 6 shows examples of the scoring points and criteria for the researcher

designed measure.

31
Table 6
Researcher Designed Outcome Measures and Scoring Approaches with Examples
Construct measured Scoring approach and examples of scoring guidelines
Problem and Solution Source:
Problem and Solution competency and Full Recall (without passage in view)
Problem and Solution Competency Score 1-8
1 = no problem, no solution, and no cause
2 = signaled cause but no problem and no solution
3 = one part of the problem and solution
4 = problem and cause but no solution or incorrect solution
5 = problem and solution (correct content of problem) and
correct content of solution
6 = problem, solution & cause of the problem mention only
in the solution part.
7 = similar to 5 but additionally presented the cause of the
problem when discussing the problem
8 = problem, solution, and cause in the problem and cause
eliminated in the solution part
Problem and Solution Full Recall Total number of ideas recalled
Comparison Text Structure Source:
(1) Signaling word scores based on fill in the blanks- Cloze Task, (2) Main idea score- Write a main idea
for the passage (with passage in view) (3) Comparison competence and full recall score- Write a recall
(without passage in view)
Signaling Word Score (Cloze Task) 7 Score 1-7
points max for each of fill-in-the-blank 1 = any word
words 2 = words that show understanding of two animals being
compared but not signaling words (e.g., “joining”)
3 = signaling words, but not for the comparison structure
(e.g., “solution”)
4 = comparison signaling words with different intent (e.g.,
“smaller than” when larger than fit the context)
5 = similar signaling word (e.g., “also like”)
6 = misspelled or parts of signaling words (e.g., “same”)
7 = exact signaling words (e.g., “same as”)
Comparison Main Idea Quality Score 1–6
1 = no mention of two ideas compared and no mention of
what attributes they were compared on
5 = correct identification of the entities compared and at least
one attribute on which they were contrasted.
6 = criteria for 5 (above) but with at least 2 attributes/issues
and one of the issues was a super-ordinate issue constructed
from the text
Adapted from Wijekumar, Meyer, Lei (2013)

Procedures

Implementation continuum of text structure intervention. The text structure

intervention is implemented through a continuum of activities grouped into five phases, see

32
Figure 1, starting with professional development and moving toward student independence.

This means that intervention activities move toward student independence but can also move

back toward professional development depending on the needs of both teachers and students.

The first phase of the continuum is professional development where teachers learn about the

intervention and practice using the intervention with upcoming lessons. The second phase is

planning for text structure. In this phase, teachers purposefully plan for the use of the text

structure strategy in upcoming lesson plans. This includes identifying the text structure of

texts to be read, finding signaling words within the texts, writing the main idea using the text

structure specific stem, as well as considering the organization of topics being covered (e.g.,

comparison, cause-and-effect, problem-and-solution). Once teachers have planned for the

use of the text structure in their lessons, they move to the teacher modeling phase. In this

phase, teachers explicitly model using the text structure strategy during classroom

instruction. Teachers model the strategy for students until they feel that students are ready

to move to the next phase: student practice with teacher guidance. In this phase, students

practice using the text structure strategy as teachers provide support either in a whole class,

group, or individual setting. This means that teachers provide support and guidance in using

the text structure strategy as students are working (e.g., pointing out signaling words,

checking use of correct main idea stem, etc.). During this phase, teachers are also checking

to see student proficiency in using the strategy. Depending on student proficiency, teachers

may go back to the modeling phase or move to the student independence phase. If students

are showing proficiency, teachers can move to the student independence phase where

students use the text structure strategy without assistance. Throughout the intervention time

frame, teachers continually return to the planning for text structure phrase as they plan for

33
upcoming lessons. Additionally, teachers can request more professional development via

modeling, coaching, and/or assistance with planning at any point during the intervention.

Text Structure Strategy Intervention

Professional Planning for Teacher Student Student
development text structure modeling practice with independence
instruction teacher
guidance

Figure 1 Implementation Continuum of Text Structure Strategy Intervention

Implementation activities. To implement the text structure strategy in middle-school

science classrooms, nine activities that moved both teachers and students through the phases

of the implementation continuum were planned for use in and out of the classroom. Once

the intervention was in progress, modifications were made to provide the greatest amount of

support for teachers. Table A-3 in the Appendix presents the timeline and activities related

to the intervention delivery and data collection.

Planned Activities and Modified Activities

1. Pre-intervention professional development (Professional development phase)

Planned activity: Two days of practice-based professional development.

• Day one: Teachers learn the research behind the text structure strategy.

• Day two: Teachers participate in multiple opportunities to practice implementing

the strategy in their classrooms.

o Review upcoming lessons and identify text structure of supporting texts

(e.g., Comparing body systems)

34
 o Identify structure-specific signaling words within texts (e.g., similar,

contrast)

o Create of graphic organizers using text structure

o Write main ideas of texts using the structure-specific stems (e.g., _____

and _____ were compared on _____, _____, and _____.)

o Create inference questions for texts (e.g., Why is heart health important

to the circulatory system?)

o Create elaboration activities for lesson topic (e.g., Research health issues

that can affect the nervous system)

• Addition professional development available as needed and requested (i.e.,

modeling lessons).

Modified activity: No modifications were made during implementation

2. Weekly - Pre-Instruction Activities (50 Minutes/Week) (Planning for text structure

phrase)

Planned activity: Two planning sessions per week (50 minutes/session)

• Analyze upcoming texts in terms of text structure with researcher support

o Identify text structure of texts

o Identify signaling words in texts

o Create graphic organizers using text structure

o Write main ideas of texts using the structure-specific stems

o Create inference questions for texts

o Create elaboration activities for lesson topic

o Researcher models how to present information to students

35
 o Teachers model presentation of information to their students.

Modified Activity: One planning session per week

• Teachers focus on TEKS

• Research suggested text-structure organization for covered material (e.g., animal

adaptations as cause and effect)

• Researcher created semi-structured lesson plans (Figure A-1 in Appendix)

o Found texts connected to current topics

o Adapted texts to be more clearly signaled

o Created graphic organizers using text structure

o Provided main idea written using text structure specific stem

o Created inference questions for text

3. Classroom-implementation (Teacher modeling, student practice with teacher

guidance, student independent practice phases)

a. Identifying text structure in science texts

Planned activity: In-class instruction about the signaling words

• Teacher identification of text structure of text being read

o Teacher models identification of signaling words (e.g., similar, because,

result)

o Teacher models classification of text structure based on the signaling

words (e.g., similar, different, alike, contrast = comparison text structure)

o Teacher gradually releases students to independence

Modified activities: In-class instruction about signaling words

• Limited practice prevented teacher release to independence

36
b. Reading using the text structure strategy with science texts

Planned activity: In-class instruction on using text structure to support reading

comprehension

• Teacher explicit instruction using text structure to support reading

comprehension

o Think-alouds and text annotation focused on comparisons, causes-and-

effects, and problems-and-solutions (e.g., “I see the word because and I

know that means the cause of an event is going to be stated after that

word”, circling signaling words and writing the text structure that goes

with the word)

o Teacher gradually releases students to independence

Modified activity: In-class instruction using semi-scripted lesson plan

• No modifications needed if teacher followed semi-scripted plan

c. Selecting important ideas from the text

Planned activity: In-class instruction on getting the gist of the text

• Instruction on using the main idea pattern to find the important ideas –

comparison (e.g., ____ and ___ were compared on ____, ____, and ____) (i.e.,

Cells, tissues, organs, and organ systems were compared on structure, function,

number of cells, and complexity)

Modified activity: In-class instruction on getting the gist using semi-scripted lesson plan

• No modifications needed when teacher followed semi-scripted plan

d. Connecting the ideas using the text structures to generate hierarchical and logical

memory structures

37
 Planned activity: In-class instruction on generating a memory tree to check mental

organization of information

Modified activity: Due to limited instructional time this activity was often left out of

lessons.

e. Generating inferences based on the diagrams and text using text structures

Planned activity: In-class instruction on using text and previous experiences and/or

additional information in the text to generate inferences, no modifications were needed

when teacher followed semi-scripted lesson plan

f. Elaborating and background knowledge extension

Planned activity: Investigation and research extending background knowledge and

motivation toward current topic

Modified activity: Due to limited instructional time no elaboration or background

knowledge extension activities were performed.

g. Monitoring reading comprehension self-check

Planned Activity: In-class support of students understanding using the text structure

strategy, no modifications were needed when teacher followed semi-scripted lesson plan

Testing. Students were randomly assigned to take either Form A or B of the GSRT. The

testing was conducted during science classes and was administered by me. The SKA was

administered by the school to all grade science students. All students at the middle school

received the text structure strategy intervention. After 10 weeks of the text structure strategy

intervention, students were given an immediate posttest. The posttest had the same condition

as the pretest.

38
 Scoring. The prose analysis system of Meyer (1975, 1985) programmed into a

computerized scoring system was be used to score the experimenter designed measures. This

approach is efficient and provides consistent high reliability to the scores.

Research Questions

1. Were there pre-posttest improvements in science and reading comprehension scores

for seventh-grade students using the text structure strategy for 10 weeks?

2. Were there differences in improvements based on gender, ethnicity, economic

status, language proficiency?

3. Were there any teacher effects on student learning outcomes?

4. Do students with lower pretest science scores (i.e., raw score below 12, 41%)

demonstrate better improvements after intervention than those with higher pretest

scores?

Method

To answer RQ1, a series of paired t-tests were conducted comparing all posttest and

pretest scores. If posttest scores were significantly higher than the pretests scores, we would

conclude that the intervention was effective.

To answer RQ2 and RQ3, ANOVA analyses were conducted on all posttest scores.

We were investigating if gender, ethnicity, economic status, language proficiency affected

posttest scores, after controlling for pretest scores. Teacher effect was examined on science

tests only.

To answer RQ4, ANOVA analyses were conducted on science knowledge

assessment gain scores (i.e., posttest-pretest scores). We first dummy coded the science

pretest capacity based on pretest raw scores. If students scored below 12 on science

39
knowledge pretest in raw score format, they would be coded as “poor science knowledge”;

otherwise they would be coded as “good science knowledge”. Therefore, the focal

independent variable is the science pretest capacity, but meanwhile we are controlling all

demographic variables. SAS version 9.4 was used for all analyses. Missing data was deleted

listwise.

40
 CHAPTER IV

RESULTS

Research Questions

The primary research question was whether there were pretest to posttest

improvements in science and reading comprehension scores for grade 7 students using the

text structure strategy for 10 weeks. Secondary research questions were (1) whether there

were differences in improvements based on gender, ethnicity, economic status, language

proficiency (2) whether there were any teacher effects on student learning, and (3) whether

students with lower pretest science knowledge scores (i.e., raw score < 12) demonstrated

better improvements after intervention than those with higher pretest scores.

Descriptive Statistics

Table 7 summarizes pretest and posttest means and standard deviations for signaling

word knowledge, GSRT Age-equivalent score, GSRT Grade-equivalent score, SKA raw

score, main idea competency, and main idea quality. Signaling word mean at pretest was

9.32 (SD = 4.70) and was 11.38 (SD = 5.74) at posttest. GSRT age-equivalent score mean at

pretest was 11.35 (SD = 2.72) and was 11.56 (SD = 2.82) at posttest. GSRT grade-equivalent

score mean at pretest was 5.60 (SD = 2.70) and was 5.81 (SD = 2.81) at posttest. SKA raw

score mean at pretest was 14.38 (SD = 4.52) and was 17.35 (SD = 2.81) at posttest. Main

idea competency mean at pretest was 1.21 (SD = 0.45) and was 1.20 (SD = 0.62) at posttest.

Main idea quality mean at pretest was 1.19 (SD = 0.43) and was 1.29 (SD = 0.51) at posttest.

Except for the main idea competency, other posttest scores are all higher than the

corresponding pretest scores. In addition, there are 126 economically disadvantaged students

41
(cf, 47 non-disadvantage), 16 students with limited English proficiency (cf. 141 non-LEP),

71 Caucasian (cf. 44 African American, 50 Hispanic, 4 Asian and 4 more than one race), 71

males (cf. 102 females). Moreover, these students are instructed by four different science

teachers (teacher A, N=14; teacher B, N=54; teacher C, N=76; teacher D, N= 29)

Table 7
Descriptive Statistics of Pretest and Posttest Scores
Pretest Posttest

N Mean SD N Mean SD
Signaling word 72 9.32 4.70 160 11.38 5.74
GSRT Age-equivalent score 88 11.35 2.72 170 11.56 2.82
GSRT Grade-equivalent score 88 5.60 2.70 170 5.81 2.81
SKA raw score 146 14.38 4.52 168 17.35 7.42
Main Idea Competency 62 1.21 0.45 155 1.20 0.62
Main Idea Quality 63 1.19 0.43 155 1.29 0.51
Note. GSRT = Gray Silent Reading Test; SKA= Science Knowledge Assessment

Text Structure Intervention Effectiveness (Primary Research Question)

Paired t-tests were conducted to compare signaling word knowledge, GSRT age-

equivalent score, GSRT grade-equivalent score, SKA raw score, main idea competency, and

main idea quality at pretest and posttest. Results showed that all scores, except main idea

competency, significantly improved after the intervention (see Table 8). There was a

significant improvement in signaling word knowledge from pretest (M= 9.32, SD= 1.14) to

posttest (M= 11.38, SD= 5.74); t(65) = 5.20, p < .0001. There was a significant improvement

on GSRT age-equivalent score from pretest (M= 11.35, SD= 2.72) to posttest (M= 11.56,

SD= 2.82); t(85) = 4.30, p < .0001. There was a significant improvement on GSRT grade-

equivalent score from pretest (M= 5.60, SD= 2.70) to posttest (M= 5.81, SD= 2.81); t(85) =

42
4.29, p < .0001. There was a significant improvement on SKA raw score from pretest (M=

14.38, SD= 4.52) to posttest (M= 17.35, SD= 7.42), t(141) = 5.97, p < .0001. There was a

significant improvement on main idea quality from pretest (M= 1.19, SD= 0.43) to posttest

(M= 1.29, SD= 51), t(60) = 2.19, p = .038. There was no significant improvement on main

idea competency from pretest to posttest.

Table 8
Paired t-Tests Comparing Pretest vs. Posttest Scores
Mean 95% CL Mean DF t Value Pr > |t| BH adjusted
diff diff p values
Signaling word 3.21 1.98 4.45 65 5.20 <.0001*** <.0001***
GSRT Age-equivalent score 0.97 0.52 1.42 85 4.30 <.0001*** <.0001***
GSRT Grade-equivalent score 0.97 0.52 1.42 85 4.29 <.0001*** <.0001***
SKA raw score 3.11 2.08 4.14 141 5.97 <.0001*** <.0001***
Main Idea Competency 0.03 -0.18 0.25 59 0.31 0.76 0.76
Main Idea Quality 0.20 0.02 0.38 60 2.19 0.0327* 0.038*
Note. GSRT = Gray Silent Reading Test; SKA= Science Knowledge Test

Effect of Gender, Ethnicity, Economic Status, Language Proficiency, and Teacher on

Intervention Effectiveness (Research Questions 2 and 3)

ANOVA analyses were conducted on all posttest scores. ANOVA analyses are

displayed from Tables 9 to 14. After controlling for pretest scores and other demographic

variables, males outperformed females on science knowledge posttests in the raw score form

(p=.015; see Table 12). The p value adjusted by the Benjamini Hochberg (BH) method was

also significant (p adjusted= .046). After controlling for pretest and other demographic

variables, economic status significantly affected main idea quality posttest scores (p= .025).

However, the BH-adjusted p value was .127, suggesting the significance may be caused by

the multiple testing error. None of the posttests were affected by English proficiency or

43
ethnicity. Also, the science posttest scores were not affected by teacher differences. To

conclude, only gender was a significant demographic predictor, but it only affected SKA

scores in the raw score format.

Demographic effects on signaling word posttest. Signaling word pretest had a

significant effect on posttest; F= 23.77, p < .0001 (see Table 9). No other variables had a

significant effect. This means that students who performed well on the signaling word pretest

also performed well on the posttest.

Table 9
Demographic Variable Effects on Signaling Word Posttests
Type III Mean
DF F Value Pr > F
SS Square
Signal word pretest 1 613.60 613.60 23.77 <.0001
Gender 1 2.87 2.87 0.11 0.74
Ethnicity 3 29.76 9.92 0.38 0.7647
LEP 1 2.33 2.33 0.09 0.7648
Economic status 1 32.03 32.03 1.24 0.2703
Error 53 1367.89 25.81

Demographic effects on GSRT age-equivalent posttest. GSRT age-equivalent

pretest had a significant effect on posttest; F= 86.92, p < .0001 (see Table 10). No other

variables had a significant effect. This means that students with a higher GSRT age-

equivalent at pretest had a higher GSRT age-equivalent at posttest.

44
Table 10
Demographic Variable Effects on GSRT Age-equivalent Posttests
Source DF Type III SS Mean F Pr > F
Square Value
GSRT age-equivalent pretest 1 386.86 386.86 86.92 <.0001
Gender 1 0.05 0.05 0.01 0.9152
Ethnicity 3 3.37 1.12 0.25 0.8595
LEP 1 0.001 0.001 0.001 0.9865
Economic status 1 9.16 9.16 2.06 0.1558
Error 72 320.45 4.45
Note. GSRT = Gray Silent Reading Test

Demographic effects on GSRT grade-equivalent posttest. GSRT grade-

equivalent pretest had a significant effect on posttest; F= 85.07, p < .0001 (see Table 11).

No other variables had a significant effect. This means that students with a higher GSRT

grade-equivalent at pretest had a higher GSRT grade-equivalent at posttest.

Table 11
Demographic Variable Effects on the GSRT Grade-equivalent Posttest
Source DF Type III Mean F Pr > F
SS Square Value
GSRT grade equivalent pretest 1 378.035 378.035 85.07 <.0001
Gender 1 0.09503 0.09503 0.02 0.8841
Ethnicity 3 3.33625 1.11208 0.25 0.8609
LEP 1 0.001 0.001 0.001 0.9868
Economic status 1 8.82042 8.82042 1.98 0.1632
Error 72 319.96 4.44
Note. GSRT = Gray Silent Reading Test

45
 Demographic and teacher effects on SKA raw score posttest. SKA raw score

pretest had a significant effect on posttest; F= 58.24, p < .0001 (see Table 12). Gender also

had a significant effect on SKA raw score posttest; F= 6.06, p = .015. No other variables

had a significant effect. This means that students with a higher SKA raw score at pretest had

a higher SKAW raw score at posttest. This also means that male students performed higher

than female students.

Table 12
Demographic Variable Effects on the SKA Posttest Raw Score
Source DF Type III SS Mean Square F Value Pr > F

SKA pretest raw score 1 2315.42 2315.42 58.24 <.0001

Gender 1 240.77 240.77 6.06 0.0153*
Ethnicity 4 118.32 29.58 0.74 0.5639
LEP 1 27.52 27.52 0.69 0.4071
Economic status 1 0.001 0.001 0.001 0.9945
Teacher 3 79.51 26.50 0.67 0.5742
Error 119 4731.29 39.76
Note. SKA= Science Knowledge Test

Demographic effects on main idea competency posttest. Main idea competency

pretest did not have a significant impact on posttest, neither did any demographic variables.

(see Table 13).

46
Table 13
Demographic Variable Effects on Main Idea Competency Posttest Scores
Source DF Type III Mean F Value Pr > F
SS Square
Main idea competency pretest scores 1 0.20 0.20 0.45 0.5058

Gender 1 0.85 0.85 1.9 0.1741

Ethnicity 3 0.34 0.11 0.26 0.8563

LEP 1 0.16 0.16 0.36 0.5496

Economic status 1 0.24 0.24 0.54 0.4664

Error 47 20.88 .44

Demographic effects on main idea quality posttest. Economic status had a

significant effect on main idea quality at posttest; F= 5.32, p = .025 (see Table 14). No other

variables had a significant effect. This means that students with from a higher economic

background performed higher at posttest.

Table 14
Demographic Variable Effects on Main Idea Quality Posttest Scores
Source DF Type III SS Mean F Value Pr > F
Square
Main idea quality pretest scores 1 0.074 0.074 0.23 0.631

Gender 1 0.64 0.641 2.02 0.162

Ethnicity 3 0.05 0.017 0.05 0.984

LEP 1 0.14 0.14 0.44 0.508

Economic status 1 1.69 1.69 5.32 0.025*

Error 48 15.23 0.32

47
Effect of Science Knowledge Pretest on Learning Outcomes (Research Question 4)

Our analyses on science gain scores suggest that the gains in science knowledge from

pretest to posttests were not significantly affected by pretest levels (i.e., those who scored

below vs. above 12 on SKA pretest raw scores; see Table 15).

Table 15
Effect of Science Pretest Capacity on Gain Scores (Science Knowledge Raw Scores)
Source DF Type III SS Mean Square F Value Pr > F

Science pretest ability 1 0.83 0.83 0.02 0.8850

Gender 1 240.74 240.74 6.06 0.0153

Ethnicity 4 120.21 30.05 0.76 0.5560

LEP 1 26.48 26.48 0.67 0.4161

Economic status 1 0.03 0.03 0.03 0.9771

Teacher 3 77.23 25.74 0.65 0.5860

Error 119 4730.88 39.76

Text Structure Strategy Use During Classroom Observations

Table 16 presents the number of observed lessons, the percentage of lessons where

text structure was appropriate, the percentage of lessons where the teacher used text

structure, the percentage of lessons I co-taught (i.e., I provided impromptu text structure

instruction in support of the planned lesson), and the percentage of lessons that I modeled

(i.e., I taught the planned lesson). Teacher A was observed 19 times. Eighty-four percent of

Teacher A’s observed lessons were appropriate for using the text structure strategy. Teacher

48
A used the text structure strategy in 32% of observed lessons and I used the text structure

strategy via co-teaching in 32% of Teacher A’s observed lessons. I modeled text structure

instruction during 37% of Teacher A’s observed lessons. Teacher B was observed 16 times.

Ninety-four percent of Teacher B’s observed lessons were appropriate for using the text

structure strategy. Teacher B used the text structure strategy in 13% of observed lessons and

I co-taught during 88% of lessons. I modeled no lessons for Teacher B. Teacher C was

observed 20 times. Ninety-percent of Teacher C’s observed lessons were appropriate for

using the text structure strategy. Teacher C used the text structure strategy in 20% of

observed lessons and I co-taught 45% of Teacher C’s observed lessons. I modeled the lesson

for 30% of Teacher C’s observed lessons. Teacher D was observed 16 times. One-hundred

percent of Teacher D’s observed lessons were appropriate for using the text structure

strategy. Teacher D used the text structure strategy in 81% of observed lessons and I co-

taught 50% of Teacher D’s lessons. I modeled the lesson for 44% of Teacher D’s observed

lessons.

Table 16
Text Structure Strategy Use During Classroom Observations
% of lessons % of lessons
where text that teacher % of lessons % of lessons
Lessons structure was used text co-taught by modeled by
Teacher observed appropriate structure researcher researcher
A 19 84.21 31.58 31.58 36.84
B 16 93.75 12.50 87.50 0.00
C 20 90.00 20.00 45.00 30.00
D 16 100 81.25 50.00 43.75
Total 71 91.55 35.21 80 28.17

49
 CHAPTER V

CONCLUSIONS

This study investigated the efficacy of the text structure strategy intervention as a

means of improving science knowledge and science reading comprehension in grade 7

science classes. The text structure strategy is a reading comprehension strategy designed to

teach students how to use the macro-structure of a text to improve reading comprehension

of science texts by selecting key ideas, stating the gist of the text (e.g., the main idea),

summarizing, making inferences, elaborating, and monitoring their reading comprehension.

The results showed that signaling word knowledge, GSRT Age equivalent score, GSRT

Grade equivalent score, the SKA raw score, and main idea quality all significantly increased

from pretest to posttest. Additionally, the results also showed that gender had a significant

effect on SKA posttest and economic status had a significant impact on main idea quality at

posttest.

Research Finding in Context

The text structure strategy is only truly useful to students if they are able to transfer

their knowledge of the strategy to various assessments and use that knowledge to improve

reading comprehension. The GSRT is a standardized measure of reading comprehension that

uses each text structure (cause-and-effect, problem-and-solution, etc.) a minimum of one

time. As a result, the GSRT is an appropriate distal measure of text structure knowledge.

The SKA can also be considered a distal measure because students must transfer their text

structure knowledge to help them comprehend and then answer science knowledge

questions. The main idea writing task is a proximal task as is the signaling word task.

However, the signaling word task is slightly more distal because students were taught to

50
identify signaling words during classroom instruction but completed a signaling word cloze

task on the research developed measure.

Text Structure Intervention Effectiveness (Primary Research Question)

The primary research question of this study was whether the text structure

intervention in the science classroom effectively improved signaling word, reading

comprehension, students’ science knowledge and main idea capacities. Paired t-tests showed

significant improvements from pretest to posttest for the GSRT, SKA, main idea quality,

and signaling word suggesting that students’ capacities in these areas improved.

In contrast, Seifert and Espin (2012) tested reading comprehension using 10 multiple

choice researcher developed questions given after each intervention session and did not find

significant differences for reading comprehension. This comparison is important because in

this study the GSRT, SKA, main idea quality, and signaling word measures were used as

measures of reading comprehension. The GSRT is a standardized measure of reading

comprehension. The SKA, main idea quality, and signaling words can also be considered

measures of reading comprehension because reading comprehension is required to answer

the science knowledge questions, write a strong main idea, and determine the best word to

complete the blank. Seifert and Espin’s intervention focused on reading fluency and

vocabulary, and they did not find effects on reading comprehension. However, this study

provided focused reading comprehension strategy instruction and found significant

differences on several measures. This difference is important because reading

comprehension in the content areas cannot be hoped for as a secondary result of instruction

on prerequisite skills such as fluency and vocabulary. The specific aim must be reading

comprehension strategies.

51
 The significant increase in main idea quality and signaling word knowledge provides

support for the use of the text structure strategy in a science classroom. Being able to state

the main idea is an important indicator of the ability to select important ideas while reading

and also take advantage of the logical connections placed in the text by the author.

Additionally, being able to correctly add signaling words to a text is an important indicator

of the ability to understand how authors connect ideas within texts.

These findings are especially significant given the modifications to teacher planning

for text structure and in-class instruction that were made and resulted in reduced dosage of

the intervention. Prior to implementation, it was planned for teachers to plan for and practice

use of the text structure strategy during the PLC time. Additionally, it was assumed that

students would be regularly reading during the science class period and have frequent

opportunities to use the strategy. However, the structure of the PLC time and limited text

resources resulted in limited teacher planning for the text structure strategy and limited in-

class instruction of the text structure strategy. As a result, I created six semi-scripted text

structure strategy lesson plans using upcoming science content for the teachers to counteract

the reduced planning time. The semi-scripted lessons were the students’ main instruction in

the text structure strategy.

Additionally, I modeled lessons for teachers A, C, and D. Teacher B’s instructional

schedule did not allow for modeling (see Table 16 for percentages). In addition to modeling,

I also co-taught with the teachers. When co-teaching, I pointed out how text structure was

useful to the current instructional activity (e.g., Test review questions used cause and effect

signaling words that could help students better understand what the question was asking)

and encouraged students to use the text structure strategy.

52
 Although I modeled, co-taught, and added general text structure support to lessons,

I observed that, except for Teacher D, the teachers were not taking full ownership of the

intervention. This can be seen in the low percentage (13-32%) of Teachers A, B, and C’s use

of the text structure strategy during observations compared with the percentage of lessons

where the text structure strategy was appropriate (84-100%). The low percentage of Teacher

A and C using the text structure strategy in during observed lessons compared with the

percentage of modeled lessons also indicates that the teachers did not full ownership of the

intervention in their classrooms. Teachers saw the text structure fully modeled and used the

semi-scripted lessons but did not actively use the text structure strategy with other materials.

The improvements in main idea quality and signaling words suggest that even in limited

doses instruction in the text structure strategy is powerful.

Effect of Gender, Ethnicity, Economic Status, Language Proficiency, and Teacher on

Intervention Effectiveness (Research Questions 2 and 3)

Additionally, this study showed that economic status had a significant impact on

main idea quality. Previous studies of science interventions and science achievement (Fang

& Wei, 2010; O’ Reilly & McNamra, 2007) have included students from various economic

backgrounds in their studies but did not report results based on economic status. This finding

goes beyond previous studies and presents an area for future investigation. It is important to

understand why and how students from economically disadvantaged backgrounds are

struggling. Future research may focus on the vocabulary knowledge and possible reading

challenges of students from economically disadvantaged background. All of the students

with limited English proficiency in this study were also from economically disadvantaged

53
backgrounds. Therefore, future research may also focus on students with limited English

proficiency.

This study found no significant differences on student performance at the teacher

level. A multi-level model analysis with students nested in classrooms was run and no effects

were found. This finding is important because teachers A, B, C, and D varied greatly in

classroom management, instructional style, and student make up. The researcher as co-

teacher model has been employed by other studies and found to be effective (Boardman et

al., 2015; Feng & Wei, 2010). However, this type of researcher-co-teacher model works well

for small scale studies but is not sustainable in larger studies. Future studies need to explore

how instructional ownership can be effectively transferred to the teacher so that they are able

to independently deliver the intervention.

The significant difference from pretest to posttest on the SKA is similar to other

findings about reading based science interventions (Cervetti et al., 2012; Fang & Wei, 2010;

Vaughn et al, 2009; Vaughn et al, 2017). However, due to the lack of a comparison group,

this study is unable to determine how much of this growth can be attributed to instruction in

the text structure strategy. The finding of significant gender effects on science knowledge is

consistent with other research on science achievement (Mau & Lynn, 2000; O’Reilly &

McNamara, 2007; Reis & Park, 2001) as well as national and state-wide results of science

achievement (NAEP, 2015; TEA, 2018). These differences may be a result of different

subject area interests. Cunningham, Hoyer, and Sparks (2015) report a significant difference

in male and female students’ interest in science. Cunningham at al. (2015) also report a

significant difference for interest in science by gender and race/ethnicity. However, the

difference may be a result of the question content. Based on the High School Transcript

54
Study, a significantly higher number of female students earned credits in advanced biology

(e.g., AP/IB biology, physiology, anatomy, and genetics) and chemistry, while a

significantly higher number of males earned credits in physics (Cunningham, 2015). The

differences in performance may then also be a result of content area knowledge and test

question content. Future research with the text structure strategy in science should

investigate why and how male and female students differ on the SKA. Based on previous

research, areas of focus may include questions about interest in science, future educational

and/or career plans, and analysis of test question content.

Effect of Science Knowledge Pretest on Learning Outcomes (Research Question 4)

This study found that student gains in science knowledge from pretest to posttest

were not significantly affected by science pretest level. This finding is significant because

it supports the text structure strategy as an equally effective intervention for all students.

The majority of students with learning disabilities spend 80% of their time in the general

education classroom, most frequently science and social studies classes (Aud et al., 2010).

A reading comprehension intervention that helps both high and low achieving students is

then perfectly suited for the science classroom where students of all abilities are expected

to learn.

Limitations of the study

This study has several limitations. This study explores the relationships among many

important factors associated with middle-grade science reading comprehension. Thus, it is

not fully powered to draw definitive causal conclusions but to serve as a starting point for a

longer research agenda. The results of this study do not generalize to a larger population of

learners. We present demographic information so that the readers may understand the

55
context for the study and research outcomes. Additionally, the study is a short duration of

10 weeks and further investigations will be necessary to study longer-term impacts of the

intervention. Further, this research uses a quasi-experimental pretest-posttest design without

a comparison group limiting the ability to draw causal conclusions from the study. Future

studies will include a comparison group and use matching techniques to compare the results

of the intervention classrooms with the comparison group classrooms on pretests and

posttests.

Significance of the Study

This study contributes to the limited body of knowledge about interventions to

improve science text reading comprehension within the middle-grade science classroom.

Many reading intervention studies focus on reading comprehension through expository and

content area texts but do so through the ELA classroom (Denton et al., 2017; Simmons et

al., 2014; Solis, Vaughn, & Scammacca, 2015; Vaughn et al., 2012; Vaughn et al., 2015;

Vaughn, Solis, Miciak, Taylor, & Fletcher, 2016; Wanzek et al., 2017). This study also adds

to the literature on the relationship between reading comprehension, as measured by a

standardized measure, main idea writing, and signaling word knowledge, and science

knowledge, as measured by a standardized test. Further, this study is significant in its ability

to help students better understand the social practices of science (i.e., sharing information

and ideas via written language) and that science texts are open to interpretation, yet textual

elements within the texts constrain the interpretations that can be made (Hand et al., 2003).

Finally, this study adds support to the idea that separate science and reading comprehension

instruction needs to traded for instruction that joins the two because science knowledge

without reading comprehension will continue to yield students who are unable to pass state

56
and national science achievement tests and ultimately be unable to live scientifically literate

lives.

57
 REFERENCES

Anderson, C.W. (1999). Inscriptions and science learning, Journal of Research in Science

Teaching, 36, 973-974.

Aud, S., Hussar, W., Planty, M., Snyder, T., Bianco, K., Fox, M., . . .Drake, L. (2010). The

condition of education 2010 (NCES 2010–028). Washington, DC: National Center

for Education Statistics, Institute of Education Sciences, U.S. Department of

Education.

Baker, L., DeWyngaert, L.U., & Zeliger-Kandasamy, A. (2015). Metacognition in

Comprehension Instruction. In S.R. Parris & K. Headley (Eds.) Comprehension

Instruction: Research-based best practices 3rd edition (p.72-87), New York, NY:

The Guildford Press.

Berkman, N.D., DeWalt, D.A., Pignone, M.P., Sheridan, S.L., Lohr, K.N, Lux, L., Sutton,

S.F. … Bonito, A.J., (2004). Literacy and Health Outcomes. (Evidence

Report/Technology Assessment No. 87). Rockville, MD: Agency for Healthcare

Research and Quality.

Boardman, A.G., Klingner, J.K., Buckley, P., Annamma, S., & Lasser, C.J. (2015). The

efficacy of Collaborative Strategic Reading in middle school science and social

studies classes. Reading and Writing, 28, 1257-1283.

Cantrell, S.C., Almasi, J.F., Carter, J.C., Rintamaa, M., & Madden, A. (2010). The impact

of a strategy-based intervention on the comprehension and strategy use of

struggling adolescent readers. Journal of Educational Psychology, 102, 257-280.

58
Cartwright, K.B. (2015). Executive function and reading comprehension: The critical role

of cognitive flexibility. In S.R. Parris & K. Headley (Eds.) Comprehension

Instruction: Research-based best practices 3rd edition (p.56-71), New York, NY:

The Guilford Press.

Cervetti, G.N., Barber, J., Dorph, R., Pearson, P.D., & Goldschmidt, P.G. (2012). The

impact of an integrated approach to science and literacy in elementary school

classrooms. Journal of Research in Science Teaching, 49, 631-658.

Chall, J. S. (1983). Stages of reading development. New York: McGraw-Hill.

Cunningham, B.C., Hoyer, K.M., & Sparks, D. (2015). Gender differences in science,

technology, engineering, and mathematics (STEM) interest, credits earned, and

NAEP performance in the 12th grade (NCES 2015-075). Washington D.C.:

American Institutes for Research.

Denton, C.A., York, M.J., Francis, D.J., Haring, C., Ahmed, Y., & Bidulescu (2017). An

investigation of an intervention to promote inference generation by adolescent poor

comprehenders. Learning Disabilities Research & Practice, 32(2), 85-98.

Fang, Z. & Wei, Y. (2010). Improving middle school students’ science literacy through

reading infusion. The Journal of Educational Research, 103, 262-273.

Farrall, M. L. (2012). Reading assessment: Linking language, literacy, and cognition.

Hoboken, NJ: John Wiley & Sons.

Foorman, B.R., Petscher, Y., Stanley, C., & Trunckenmiller, A. (2016). Latent profiles of

reading and language and their association with standardized reading outcomes in

59
 kindergarten through tenth grade. Journal of Research on Educational

Effectiveness, 107, 884-899, DOI: 10.1080/19345747.2016.1237597

Gersten, R., Fuchs, L.S., Williams, J.P. & Baker, S. (2001). Teaching reading

comprehension strategies to students with learning disabilities: A review of

research. Review of Educational Research, 71, 279-320.

Hand, B.M., Alvermann, D.E., Gee, J., Guzzetti, B.J., Norris, S.P., Phillips, L.M., Prain,

V., & Yore, L.D. (2003). Message from the “Island Group”: What is literacy in

science literacy?. Journal of Research in Science Teaching, 40, 607-615.

Harris, K.R., Lane, K.L., Graham, S., Driscoll, S.A., Sandmel, K., Brindle, M., &

Schatschneider, C. (2012). Practice-based professional development for self-

regulated strategies development in writing: A randomized controlled study.

Journal of Teacher Education, 63, 103-119.

Kaldenberg, E.R., Watt, S.J., & Therrien, W.J., (2015). Reading instruction in science for

students with learning disabilities: A meta-analysis. Learning Disability Quarterly,

38, 160-173.

Kukner, J.M., & Orr, A.M. (2015). Inquiring in to pre-service content area teachers’

development of literacy practices and pedagogical content knowledge. Australian

Journal of Teacher Education, 40(5), 41-60.

Lemke, J. (1990). Talking science: Language, learning, and values. Norword, NJ: Ablex

Publishing Corporation

60
Mau, W. & Lynn, R. (2000). Gender differences in homework and test scores in

mathematics, reading and science at tenth and twelfth grade. Psychology,

Evaluation & Gender, 2, 119-125.

Meyer, B. J. F. (1975). The organization of prose and its effects on memory.

Amsterdam: North -Holland.

Meyer, B. J. F., Brandt, D. M., & Bluth, G. J. (1980). Use of the top-level structure in text:

Key for reading comprehension of ninth-grade students. Reading Research

Quarterly, 16, 72-103.

Meyer, B.J.F., & Ray, M.N. (2011). Structure strategy interventions: Increasing reading

comprehension of expository text. International Electronic Journal of Elementary

Education, 4, 127-152.

Meyey, B.J.F., Middlemiss, W., Theodorou, E., Brezinski, K.L., & McDougall, J. (2002).

Effects of structure strategy instruction delivered to fifth-grade children using the

internet with and without the aid of older adult tutors. Journal of Educational

Psychology. 94, 486-519.

Meyer, B.J.F., Wijekumar, K., Middlemiss, W., Higley, K., Lei, P., Meier, C., Spielvogel,

J. (2010). Web-based tutoring of the structure strategy with or without elaborated

feedback or choice for fifth- and seventh-grade readers. Reading Research

Quarterly, 41, 62-92.

Meyer, B.J.F., & Wijekumar, K. (2014). Why fifth- and seventh-graders submit off-task

responses to a web-based reading comprehension tutor rather than expected

learning responses. Computers & Education, 75, 229-252.

61
National Assessment of Educational Progress (NAEP) 2015. Available at

http://www.nationsreportcard.gov/reading_math_2015/#reading?grade=4 on April

27, 2017

Norris, S. P. & Phillips, L.M. (2003), How literacy in its fundamental sense is central to

scientific literacy. Science Education, 87, 224-240.

O’Reilly, T. & McNamara, D.S. (2007). The impact of science knowledge, reading skill,

and reading strategy knowledge on more traditional “high-stakes” measures of high

school students’ science achievement. American Educational Research Journal, 44,

161-196.

Pressley, M., & Wharton-McDonald, R. (1997). Skilled comprehension and its

development through instruction. School Psychology, 26, 448-466.

Reis, S. M., & Park, S. (2001). Gender differences in high-achieving students in math and

science. Journal for the Education of the Gifted, 25(1), 52-73.

Seifert, K., & Espin, C. (2012). Improving reading of science text for secondary students

with learning disabilities: Effects of text reading, vocabulary learning, and

combined approaches to instruction. Learning Disability Quarterly, 35, 236-247.

Shanahan, T., & Shanahan, C. (2008). Teaching disciplinary literacy to adolescents:

Rethinking content-area literacy. Harvard Educational Review, 78, 40-59.

Simmons, D., Fogarty, M., Oslund, L.S., Hairrell, J.D., Anderson, L., Clemens, N.,

Vaughn, S., Roberts, G., Stillman, S., & Fall, A. (2014) Integrating content

knowledge-building and student-regulated comprehension practices in secondary

62
 English language arts classes. Journal of Research on Educational Effectivenss, 7,

309-330.

Solis, M., Vaughn, S., & Scammacca, N. (2015). The effects of an intensive reading

intervention for ninth graders with very low reading comprehension. Learning

Disabilities Research & Practice, 30, 104-113.

Swanson, E., Wanzek, J., Vaughn, S., Roberts, G., & Fall, A. (2015). Improving reading

comprehension and social studies knowledge among middle school students with

disabilities. Exceptional Children, 81, 426-442.

Texas Education Agency (2011). 19 TAC Chapter 112. Texas Essential Knowledge and

Skills for Science. Retrieved from

http://ritter.tea.state.tx.us/rules/tac/chapter112/index.html

Texas Education Agency (2017). State of Texas Assessments of Academic Readiness

(STAAR) Performance Labels and Policy Definitions. Retrieved from

https://tea.texas.gov/student.assessment/staar/performance-standards/

Texas Education Agency (2018). Texas Academic Performance Reports. Retrieved from

https://tea.texas.gov/perfreport/tapr/index.html

Vaughn, S., Martinez, L.R., Linan-Thompson, S., Reutebuch, C.K. Carlson, C.D., &

Francis, D.J. (2009). Enhancing social studies vocabulary and comprehension for

seventh-grade English language learners: Findings from two experimental studies.

Journal of Research on Educational Effectiveness, 2, 297-324.

63
Vaughn, S., Martinez, L.R., Wanzek, J., Roberts, G., Swanson, E., & Fall, A. (2017).

Improving content knowledge and comprehension for English language learners:

Findings from a randomized control trial. Journal of Educational Psychology,

109(1), 22-34.

Vaughn, S., Roberts, G., Wexler, J., Vaughn, M.G., Fall, A., & Schnakenberg, J.B. (2015).

High school student with reading comprehension difficulties: Results of a

randomized control trial of a two-year reading intervention. Journal of Learning

Disabilities, 48, 546-558.

Vaughn, S., Solis, M., Miciak, J., Taylor, W.P., & Fletcher, J.M. (2016). Effects from a

randomized control trial comparing researcher and school-implemented treatments

with fourth graders with significant reading difficulties. Journal of Research on

Educational Effectiveness, 9, 23-44.

Vaughn, S., Wexler, J., Leroux, A., Roberts, G., Denton, C., Barth, A., & Fletcher, J.

(2012). Effects of intensive reading intervention for eighth-grade students with

persistently inadequate response to intervention. Journal of Learning Disabilities,

45, 515-525.

Waznek, J., Petscher, Y., Al Oltaiba, S., Rivas, B.K., Jones, F.G., Kent, S.C.,

Schatschneider, C., & Mehta, P. (2017). Effects of a year long supplemental

reading intervention for students with reading difficulties in fourth grade. Journal

of Educational Psychology, 109, 1103-1119.

Wijekumar, K., Meyer, B.J.F., Lei, P., Lin, Y., Johnson, L.A., Shurmatz, K., Spielvogel, J.,

Ray, M.N., & Cook, M. (2014). Improving reading comprehension for 5th grade

64
 readers in rural and suburban schools using web-based intelligent tutoring systems.

Journal of Research on Educational Effectiveness. 7, 331-357.

Wijekumar, K., Meyer, B.J.F., Lei, P. (2012). Large-scale randomized controlled trial with

4th graders using intelligent tutoring of the structure strategy to improve nonfiction

reading comprehension. Journal of Educational Technology Research and

Development. 60, 987-1013

Wijekumar, K., Meyer, B.J.F., Lei, P.-W. (2013). High-fidelity implementation of web-

based intelligent tutoring system improves fourth and fifth graders content area

reading comprehension. Computers & Education, 68, 366-379.

Wijekumar, K. (K.), Meyer, B. J. F., & Lei, P. (2017). Web-based text structure strategy

instruction improves seventh graders’ content area reading comprehension. Journal

of Educational Psychology, Advance online publication.

http://dx.doi.org/10.1037/edu0000168.

Yore, L. (1991). Secondary science teachers’ attitudes toward and beliefs about science

reading and science textbooks. Journal of Research in Science Teaching, 28, 55-72.

Yore, L., Bisanz, G.L., & Hand, B.M. (2003). Examining the literacy component of

science literacy: 25 years of language arts and science research. International

Journal of Science Education, 25, 689-725.

65
 APPENDIX

Table A-1
Studies in literature review
Study Research design Measures Duration and Results
and population Conditions
O’Reilly & Exploratory • Multiple-choice science N/A Science knowledge on:
McNamara (2007) analysis knowledge test (α=.74) -Multiple choice d=0.51
• Gates-MacGinitie(α=.95) -Open-Ended Comprehension, d=0.58
Grades 9-12 • Metacomprehension Strategy -Course Grade, d=0.20
Index (α=.72) -State Science Test, d=0.71
n=1651 students • Science comprehension passage
(Open ended questions, α=.87, Reading Skill on:
multiple-choice questions, -Multiple choice d=0.64
α=.83) -Open-Ended Comprehension, d=0.78
-Course Grade, d=0.25
-State Science Test, d=0.73

Comparison of skilled readers with low

science knowledge and less skilled readers
with higher science knowledge on:
-Multiple choice d=0.14
-Open-ended passage comprehension,
d=0.30

Reading skill for high knowledge students:

-Multiple-choice d=0.94
-Open-ended d=1.01
-Grade d=.43
-State science d=0.81

66
Table A-1 (continued)
Study Research design Measures Duration and Results
and population Conditions
Seifert & Espin Quasi-experimental -Reading fluency: 5 min to read Duration: Text-reading on:
(2012) Grade 9-12 passage aloud 4 sessions, one per -Reading fluency ES=0.47 (compared to
n=20 students with -Vocabulary knowledge: 5 min for condition Vocabulary condition), ES= 0.97
LD vocabulary matching task 45 min/session (compared to control)
-comprehension measure: 10 multiple Conditions:
choice questions -Text-reading Combined condition on:
-Vocabulary knowledge -Reading fluency ES=0.49 (compared to
-Text-reading and Vocabular condition), ES=1.04 (compared
vocabulary knowledge to control condition)
-Control
Vocabulary condition on:
-Reading fluency ES=0.53 (compared to
control)

Cervetti, Barer, Classroom based -Science understanding Duration Science understanding d=.65
Drop, Pearson, & random-assignment (α=.84/pretest, α=.81/posttest) 40 sessions (4 Science vocabulary d=.23
Goldschmidt (2012) Grade 4 -Science writing (inter-rater investigations at 10 Science writing:
n=94 teachers reliability= .85/pretest, .79/posttest sessions each) -overall d=.40
-Science vocabulary (α=.46/pretest, Conditions: -science concepts: d=.63
.69/posttest -Research developed -vocabulary count d=.80
-Reading comprehension integrated science- -evidence d=.33
(α=.77/pretest, .76/posttest) literacy unit on topic of -introduction d=.38
light -clarity d=.43
-Regularly used
curriculum on topic of
light
Fang & Wei (2010) Quasi-experimental -Gates-MacGinitie Reading Duration: ISR on:
Grade 6 Test(GMRT) 22 weeks -GMRT vocabulary d=.23
n=2 teacher -curriculum-referenced science test Conditions: -GMRT comprehension d=.22
n=233 students (CRST) (α=.78) -Inquiry-based science -GMRT total score d=.22
-academic year science grade(AYSG) curriculum with reading -CRST d=.35
(ISR) -AYSG d=.34
-Inquiry based science

67
Table A-1 (continued)
Study Research design Measures Duration and Results
and population Conditions
Boardman, Multi-site cluster - Gates-MacGinitie Reading Duration: -GMRT comprehension g=.18 (full), g=.13
Klingner, Buckley, randomized trial Test(GMRT) School year (partial)
Annamma, & Grades 6-8 - Reading and writing scale scores Conditions: -State reading test g=.19 (full), g=.14
Lasser (2015) n=9 teachers from state mandated test -Collaborative Strategic (partial)
n=1074 students Reading (CSR) (full) -State writing test g=.23 (full), g=.15
-Collaborative Strategic (partial)
Reading (CSR) (partial)
-Business as usual
Kukner & Orr Qualitative -semi-structured interviews N/A Themes associated with PSTs with strong
(2015) n=16 preservice commitment to literacy:
content area -Expanded understandings of literacies
teachers (PST) -literacy routines as opportunities for
thinking and learning
-Clear connections to curriculum outcomes
and relevant authentic assessments
Themes associated with PSTs less certain
about role of literacy in teaching:
-inability to speak fluently about
incorporating literacy into teaching
-lack of metacognitive awareness on how to
plan for literacy strategies
-lack of awareness of need to model
literacy strategies

68
Table A-1 (continued)

Study Research design Measures Duration and Results

and population Conditions
Yore (1991) Grades 6-12 -Science and Reading Questionnaire N/A Attitudes toward science reading
n=215 science instruction
teachers Strongly Agree= 7, Strongly Disagree=1
1. A science teacher is obliged to help
students improve their reading abilities
(M= 5.87, SD= 1.06)
2. Science teachers should teach content
and leave reading instruction to reading
teachers (M= 2.73, SD= 1.79)
3. Knowing how to teach reading in
science should be required for teaching
certification (M= 5.00, SD= 1.07)
4. Science teachers should be familiar
with the theoretical concepts of the
reading process (M= 5.07, SD= 1.22)
5. Only teachers of English should be
responsible for teaching reading in
secondary schools.
(M= 2.27, SD= 1.10)
6. Every science teacher should teach
students how to read science materials
(M= 5.80, SD= 0.78)
7. The primary responsibility of a science
teacher should be to impart subject
matter knowledge.
(M= 3.53, SD= 1.19)
8. A science teacher should be
responsible for helping students
comprehend at an interpretive level as
well as a literal level when they read.
(M= 5.40, SD= 0.83)

69
Table A-1 (continued)
Study Research design Measures Duration and Results
and population Conditions
Yore (1991) 9. Science teachers should help students
learn to set purposes for reading and
how to monitor their own success.
(M= 5.27, SD= 0.96)
10. Science teachers should feel a greater
responsibility to the content they teach
than to any reading instruction they
may be able to provide.
(M= 5.00, SD= 1.25)
11. Teachers who want to improve
students’ interest in reading should
show that they like to read.
(M= 5.73, SD= 0.96)
12. It is not necessary for the reader to
know the purpose for reading science
text material. (M= 2.40, SD= 1.18)
Beliefs about models of reading, science
texts, and science reading skills
Strongly Agree= 7, Strongly Disagree=1
1. The science text and the reader interact
to invent new meaning not contained in
the text or by the reader.
(M= 4.23, SD= 1.23)
2. Prior science experience is required for
science text reading to be more than
just an exercise in memorization.
(M= 4.68, SD= 1.41)
3. The reader brings the meaning to a
science text which just stimulates the
readers’ understanding.
(M= 4.13, SD= 1.00)

70
Table A-1 (continued)

Study Research design Measures Duration and Conditions Results

and population
Yore (1991) 4. While comprehending a new science
concept from reading, most readers
relate text information to familiar
examples in their memory.
(M= 5.01, SD= 1.31)
5. The science text contains all the
information needed by the reader to
understand the idea or concept.
(M= 2.39, SD= 1.21)
6. Students require no background on a
topic to read and comprehend text on
that topic. (M= 2.87, SD= 1.21)
7. Most texts for secondary science are
written at or below grade level for
which they are intended, as judged by
readability formula.
(M= 3.52, SD= 1.43)
8. Sentence and paragraph structure in
science textbooks must necessarily be
different from that of other texts.
(M= 3.73, SD= 1.58)
9. Students who can read non-science
texts at their grade level will have no
trouble reading science texts.
(M= 4.93, SD= 0.96)
10. The problem with poor readers is that
they do not follow the logical structure
of paragraphs. (M= 4.67, SD= 0.72)
11. The ability to predict upcoming text
can be used to distinguish between
good and poor readers.
(M= 4.87, SD= 0.99)

71
Table A-1 (continued)

Study Research design Measures Duration and Conditions Results

and population
Yore (1991) 12. Titles and headings in a text are useful
for effective reading comprehension.
(M= 6.27, SD= 0.59)
13. Technical vocabulary should be
introduced to students in content
classes before they meet those terms in
a reading passage.
(M= 4.40, SD= 1.55)

72
Table A-2
TEKS Used in Science Knowledge Assessment
Number
TEKS # Texas Essential Knowledge and Skills description
of
questions
(11) Earth and space. The student understands the organization of our solar system and
6.11b the relationships among the various bodies that comprise it. The student is expected to: 1
(B) understand that gravity is the force that governs the motion of our solar system;
(12) Organisms and environments. The student knows all organisms are classified into
Domains and Kingdoms. Organisms within these taxonomic groups share similar
characteristics which allow them to interact with the living and nonliving parts of their
6.12d ecosystem. The student is expected to: (D) identify the basic characteristics of organisms, 1
including prokaryotic or eukaryotic, unicellular or multicellular, autotrophic or
heterotrophic, and mode of reproduction, that further classify them in the currently
recognized Kingdoms;
Matter and energy. The student knows the differences between elements and compounds.
6.5c The student is expected to: (C) differentiate between elements and compounds on the most 1
basic level;
Matter and energy. The student knows matter has physical properties that can be used for
6.6a classification. The student is expected to: (A) compare metals, nonmetals, and metalloids 1
using physical properties such as luster, conductivity, or malleability;
Matter and energy. The student knows matter has physical properties that can be used for
6.6b classification. The student is expected to: (B) calculate density to identify an unknown 1
substance; and
Force, motion, and energy. The student knows force and motion are related to potential
6.8a and kinetic energy. The student is expected to: (A) compare and contrast potential and 2
kinetic energy;
Force, motion, and energy. The student knows force and motion are related to potential
6.8c and kinetic energy. The student is expected to: (C) calculate average speed using distance 2
and time measurements;
Force, motion, and energy. The student knows force and motion are related to potential
6.8d and kinetic energy. The student is expected to: (D) measure and graph changes in 2
motion;
Force, motion, and energy. The student knows that the Law of Conservation of Energy
states that energy can neither be created nor destroyed, it just changes form. The student is
6.9c 2
expected to: (C) demonstrate energy transformations such as energy in a flashlight battery
changes from chemical energy to electrical energy to light energy.
(10) Organisms and environments. The student knows that there is a relationship between
7.10b organisms and the environment. The student is expected to: (B) describe how biodiversity 1
contributes to the sustainability of an ecosystem;
(10) Organisms and environments. The student knows that there is a relationship between
organisms and the environment. The student is expected to: (C) observe, record, and
7.10c 1
describe the role of ecological succession such as in a microhabitat of a garden with
weeds.

73
Table A-2 (continued)
Organisms and environments. The student knows that populations and species demonstrate
variation and inherit many of their unique traits through gradual processes over many
7.11a 1
generations. The student is expected to: (A) examine organisms or their structures such as
insects or leaves and use dichotomous keys for identification;
Organisms and environments. The student knows that populations and species demonstrate
variation and inherit many of their unique traits through gradual processes over many
7.11c generations. The student is expected to: (C) identify some changes in genetic traits that 1
have occurred over several generations through natural selection and selective breeding
such as the Galapagos Medium Ground Finch (Geospiza fortis) or domestic animals.
(12) Organisms and environments. The student knows that living systems at all levels of
organization demonstrate the complementary nature of structure and function. The student
7.12b is expected to: (B) identify the main functions of the systems of the human organism, 2
including the circulatory, respiratory, skeletal, muscular, digestive, excretory,
reproductive, integumentary, nervous, and endocrine systems;
(12) Organisms and environments. The student knows that living systems at all levels of
organization demonstrate the complementary nature of structure and function. The student
7.12d is expected to: (D) differentiate between structure and function in plant and animal cell 1
organelles, including cell membrane, cell wall, nucleus, cytoplasm, mitochondrion,
chloroplast, and vacuole;
(12) Organisms and environments. The student knows that living systems at all levels of
organization demonstrate the complementary nature of structure and function. The student
7.12f is expected to: (F) recognize that according to cell theory all organisms are composed of 1
cells and cells carry on similar functions such as extracting energy from food to sustain
life.
(14) Organisms and environments. The student knows that reproduction is a characteristic
of living organisms and that the instructions for traits are governed in the genetic material.
7.14b
The student is expected to: (B) compare the results of uniform or diverse offspring from
sexual reproduction or asexual reproduction;
(14) Organisms and environments. The student knows that reproduction is a characteristic
of living organisms and that the instructions for traits are governed in the genetic material.
7.14c 1
The student is expected to: (C) recognize that inherited traits of individuals are governed
in the genetic material found in the genes within chromosomes in the nucleus.
(5) Matter and energy. The student knows that interactions occur between matter and
7.5c energy. The student is expected to: (C) diagram the flow of energy through living systems, 1
including food chains, food webs, and energy pyramids.
(6) Matter and energy. The student knows that matter has physical and chemical
properties and can undergo physical and chemical changes. The student is expected to:
7.6a 2
(A) identify that organic compounds contain carbon and other elements such as
hydrogen, oxygen, phosphorus, nitrogen, or sulfur;
(7) Force, motion, and energy. The student knows that there is a relationship among force,
motion, and energy. The student is expected to: (A) contrast situations where work is
7.7a 2
done with different amounts of force to situations where no work is done such as moving a
box with a ramp and without a ramp, or standing still;
(8) Earth and space. The student knows that natural events and human activity can impact
7.8c Earth systems. The student is expected to: (C) model the effects of human activity on 1
groundwater and surface water in a watershed.

74
Table A-3
Timeline of Intervention Activities

Date Activity

October Professional development day 1

October Professional development day 2

November 2, Pretest
2018

Intervention

Week Class Classroom Activities Intervention activities Teacher use Researcher Researcher Text
visit of text use of text modeled structure
structure structure lesson appropriate
intervention intervention lesson
1-5 -Students drew pictures -Suggestion to use comparison matrix to highlight N Y N Y
of germinated pinto differences between the beans and facilitate making
beans and answered inferences related to the differences. Teacher
questions about changes rejected idea stating students made comparisons
in the beans. visually and did not need to write them down
6 -Describe plants with -Students given intervention adapted text and N Y N Y
high and low turgor organizer to read and complete silently. Teacher did
pressure. not model for students. (previous lesson)
-Suggestion that teacher work with and model for
-Students compared students when using text structure before giving as
potatoes soaked in plain independent activity
and salt water. Used - Suggestion to use comparison matrix was rejected
“Venn diagram” because teacher “needed them to describe it”.
1 Nov27-Dec 1

-Suggestion to use matrix instead of Venn diagram.

7 -Text on turgor pressure
Helped students create matrix. Used cause and effect
language to help students see differences.
-Modeled lesson N Y Y Y
-Discussed ways to use text structure in class
-created organizer for next text

75
Table A-3 (continued)
8 -Turgor pressure potato -Teacher created comparison matrix for potatoes N Y N Y
lab based on suggestion
-Comparison words and cause and effect words used
1 Nov27-Dec 1

during lab
9 -Notes on stimulus and -Teacher used cause and effect words to make Y Y N Y
response explicit the relationship between stimulus and
response
-Supported teacher use of cause and effect to explain
stimulus and response

1-2 -Oral notes on -Teacher uses cause and effect words Y Y Y Y

homeostasis
-Showed students how to use text structure with
high stakes test questions

3 -Notes on homeostasis -Use of cause and effect words and cause and effect Y Y Y Y
chart

-Showed students how to use text structure with

high stakes test questions
2 Dec 4-8

4-5 -Notes on homeostasis -Modeled using of text structure Y Y N Y

-Showed students how to use text structure with

high stakes test questions

1-8 -Genetics -Modeled text structure lesson N Y Y Y

-Teacher used class period to grade

9, 12 -Genetics -Used cause and effect to explain Punnett squares to N Y N Y

3 Dec. 11-15

students

10- -Genetics -Modeled use of text structure intervention to help N Y Y Y

11, with genetics
13-14

76
Table A-3 (continued)

1-2, 4 -Review of school rules N N N N

and lockdown procedure

3 -Genetics quiz -Helped students identify the signaling words in quiz N Y N Y

questions and how those could help with answering
questions

5 -Organizer for -Suggested that students use a comparison matrix for N Y N Y

reproduction sexual and asexual reproduction

6 -Review of comparison -Modeling of text structure and writing main idea Y Y Y Y

activity from previous
day
4 Jan. 8-12

-Animal adaptation
notes

7 -Animal adaption notes -Modeling of text structure and writing main idea Y Y Y Y
5 Jan 16-10

No observations due to testing

1 -Warm up question -Suggestion of answering warm up question using N Y N Y

cause and effect structure
-animal adaption notes
-Suggestion of using cause and effect statements
rather than coping notes verbatim

2 -Animal adaptation -Teacher using text structure and modeling for Y N N Y

notes students

-Teacher asks students to identify the cause and the

6 Jan 22-26

effect

-Students wrote main idea using cause and effect

stem

77
Table A-3 (continued)

3, 4 -Warm up question -Worked with students on writing main idea using Y Y Y Y

cause and effect stem
-Animal adaptations
-Worked with students on using cause and effect to
make inferences

1 -Warm up question -Suggested that students answer warm up using N Y N N

cause and effect structure
-Dichotomous key
-Dichotomous keys do not support use of text
structure intervention

2-3 -Dichotomous key -Dichotomous keys do not support use of text N N N N

structure intervention

4 -warm up question -Teacher used cause and effect signaling words to Y N N N

help students with warm up question
-Dichotomous key
-Dichotomous keys do not support use of text
structure intervention

5 -dichotomous key -Dichotomous keys do not support use of text N N N N

structure intervention

6-7 -warm up question -Suggested that students use the signaling words in N Y N Y
the question to better understand what the question
-Stations about was asking
dichotomous keys
-Helped students apply text structure to the various
stations as appropriate

8 -warm up questions -warm up and notes are based on recitation of N N N N

memorized information and not appropriate for text
7 Jan 29-Feb 2

-notes on natural structure intervention

selection

9 -Finishing text from -Reminded students to use the cause and effect stem N Y N Y
previous day when writing the main idea of the passage

78
Table A-3 (continued)

10 -Dichotomous key -Dichotomous keys do not support use of text N N N N

structure intervention

1 -Test review -Explained to students how most questions were N Y N Y

written using cause and effect and if students wrote
answer using cause and effect structure they would
better remember the information

2 -warm up question -Helped students set up comparison matrix for warm N Y N Y

up question
-test review
-Explained to students how most questions were
written using cause and effect and if students wrote
answer using cause and effect structure they would
better remember the information

3 -Comparison of sexual -Teacher is not using matrix for sexual and asexual N Y N Y
and asexual comparison.
reproduction
-Explained to students how most questions were
-Test review written using cause and effect and if students wrote
answer using cause and effect structure they would
better remember the information

4, 5 -Warm up question -Explained that both warm up questions were causes N Y N Y

and effect and should be answered as such
-Test review
8 Feb 5-9

-Explained to students how most questions were

written using cause and effect and if students wrote
answer using cause and effect structure they would
better remember the information.

1, 3, 5 -Energy web -Teacher mentioned that previous lesson using Y Y N Y

comparison text went well because students
understand the format
9 Feb 12-15

-Teacher uses cause and effect signaling words to

discuss changes in food web. Students are asked to
identify the cause and the effect

79
Table A-3 (continued)

2 -Energy web -Teacher does not use signaling words to explain N N N Y

energy web.

4 -Fill in the blank notes -Students completed comparison matrix over N Y N N

producers, consumers, and decomposers
independently (previous lesson). Teacher had not
gone over comparison or given it back to students.

-Attempted to help students with notes but fill the

blank format made text structure intervention was
not easily applied
10 Feb

1-5 -Creating a food for -Encouraged students to think about how the N Y N Y
19-23

ecosystem of choice ecosystem would affect the type of animals that live
there

80
Figure A-1
Semi-scripted lesson plan
Adaptations Lesson Plan

1. Tell students that this text is cause and effect and as they read they will need to think about
what the causes are and what the effects are. Specifically, they need to think about adaptations
as a cause and what the effects of those adaptations might be.
2. Read article WITH students and point out the cause/effect signaling words (highlighted in
teacher text). Work with students to fill in graphic organizer as you read, making sure students
understand how the cause/effects are related. For example, give students the cause and ask
them to state what the effect is.
3. Discuss with students what the major cause being discussed in each section is and what the
major effect of that cause is. Have students write the main idea for each section using the cause
and effect main idea pattern (The cause is _____ and the effect is _____).
4. Discuss with students how the three causes work together to create a main effect. Have
students write the main idea for the whole passage using the cause and effect main pattern.
5. Have students answer inference questions based on information in the text and using the cause
and effect organization to help them in making those inferences.

Adaptations Graphic Organizer (Teacher copy)

CAUSE EFFECT

Behavioral Adaptations

Heron raises wings to block out sun’s glare Easier for heron to see prey
Organisms have behaviors that
Behavioral adaptations
support survival and reproduction
Fish not knowing how to swim Would probably die
Would probably die before
Wolf not learning to hunt
reproducing
Structural Adaptations

Birds have light bones Easier to fly

Fish have gills Breathe underwater

Physical characteristics that
Structural adaptations
support survival
Physiological Adaptations
Body is protected from injury and
Oyster’s ability to make shell
predators
Physiological adaptation (biochemical Organism is better suited to
function) environment
Human’s ability to maintain constant
Live in a variety of climates
internal body temperature
Structural, physiological, and behavioral
Organism’s survival
adaptations work together
Lion’s sharp teeth claws, ability to stalk
Lion is well suited to hunting large
prey, and ability to grow sharp teeth and
animals, more likely to survive
claws

81
Figure A-1 (continued)
Main Idea (Teacher Copy)
The main idea of a text tells the topics that were discussed in the text, how the topics
were discussed (text structure) and what was discussed about the topics. Write a main
idea sentence about the article using the cause and effect pattern.
Cause and Effect Main Idea Sentence Stem
The main cause is __________________, and the main effect is
_________________________.

Write your main idea sentence here:

The cause is behavioral adaptations. The effect is an organism has behaviors that help
it survive.
The cause is structural adaptation. The effect is an organism has physical
characteristics that help it survive.
The cause is physiological adaptations. The effect is an organism has biochemical
functions that help it survive.
The causes are behavioral, structural, and physiological adaptations that support an
organism’s survival. The effect is the adaptations that helped with survival are passed
on AND the population survives.

Inference questions:

1. Kangaroo rats live in the desert. Their kidneys can produce very concentrated urine
to conserve water. What is the effect of the rats’ ability to produce concentrated
urine? How do the rats’ internal structures function to help it survive its
environment?
2. How does the human behavioral adaptation of language help us to survive?
3. A snake’s venom is a physiological adaptation. How does this adaptation support
the survival of venomous snakes?
4. Why do behavioral, structural, and physiological adaptations vary across species of
animal?
5. In the far north, near the artic, there are artic hares that are tan in the spring and
white in the winter. How do you think this physiological adaptation helps the hare
survive?

82
6. A population of grasshoppers lives in a field of green grass. Some of the
grasshoppers are dark brown and some are green. Which grasshoppers are more
likely to be prey for birds?
7. What is likely to happen to the grasshopper population over several generation?

83